Commit graph

250 commits

Author SHA1 Message Date
Daniel Han
dae0870fd7 Gemma2 (#723) July-2024
* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

---------

Co-authored-by: Michael <107991372+shimmyshimmer@users.noreply.github.com>
2024-07-03 12:12:21 -07:00
Daniel Han
cfafc707a1 Gemma2 (#709)
* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* offload

* Update llama.py

* Create a starter script for command-line training to integrate in ML ops pipelines. (#623)

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* ollama

* Update mapper.py

* Update chat_templates.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Fixes

* clearer messages

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* log

* Update __init__.py

* Update llama.py

* Update __init__.py

* Create Merge.png

* Create ollama.png

* Gemma2

* Update llama.py

* Update loader.py

* Update pyproject.toml

* Update pyproject.toml

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update _utils.py

* Revert Gemma2

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update rms_layernorm.py

* Update gemma2.py

* logit softcapping

* Update cross_entropy_loss.py

* Update llama.py

* Update llama.py

* Update gemma2.py

* Update gemma2.py

* Update cross_entropy_loss.py

* Update llama.py

* Update llama.py

* Update cross_entropy_loss.py

* Update cross_entropy_loss.py

* Update llama.py

* Update cross_entropy_loss.py

* Update cross_entropy_loss.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update llama.py

* Update gemma2.py

* Update llama.py

* Update llama.py

* Update gemma2.py

* Update gemma2.py

* Update llama.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update gemma2.py

* Update _utils.py

* Update _utils.py

* Update gemma2.py

* compile flags

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update gemma2.py

* Update gemma2.py

* fixes

* Update _utils.py

* Fix generation

* Update llama.py

* Update llama.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* pad token

* Update gemma2.py

* pad token

* Update _utils.py

* Update llama.py

* Update gemma2.py

* edit warning

* Update tokenizer_utils.py

---------

Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-07-02 22:51:01 -07:00
Daniel Han
71e5bad518 Nightly (#676)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* offload

* Update llama.py

* Create a starter script for command-line training to integrate in ML ops pipelines. (#623)

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* ollama

* Update mapper.py

* Update chat_templates.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Fixes

* clearer messages

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* log

* Update __init__.py

* Update llama.py

* Update __init__.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-21 15:32:26 +10:00
Daniel Han
c19e2e1015 Nightly (#673)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* offload

* Update llama.py

* Create a starter script for command-line training to integrate in ML ops pipelines. (#623)

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* ollama

* Update mapper.py

* Update chat_templates.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Fixes

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-21 00:28:52 +10:00
Daniel Han
abaac826e1 Ollama (#671)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* offload

* Update llama.py

* Create a starter script for command-line training to integrate in ML ops pipelines. (#623)

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* ollama

* Update mapper.py

* Update chat_templates.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-20 22:28:28 +10:00
Daniel Han-Chen
14362c225a Update chat_templates.py 2024-06-20 19:49:38 +10:00
Daniel Han-Chen
b62451f912 Update chat_templates.py 2024-06-20 19:45:02 +10:00
Daniel Han
733cd5b48b Ollama bug fixes (#667)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* offload

* Update llama.py

* Create a starter script for command-line training to integrate in ML ops pipelines. (#623)

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* ollama

* Update mapper.py

* Update chat_templates.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* Update save.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-20 04:55:13 +10:00
Daniel Han
196635e959 Ollama (#665)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* offload

* Update llama.py

* Create a starter script for command-line training to integrate in ML ops pipelines. (#623)

* Update chat_templates.py

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-19 04:53:26 +10:00
Daniel Han
b9a71ac370 Fix continuing LoRA finetuning (#656)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

* Update llama.py

* peft

* patch

* Update loader.py

* retrain

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-17 00:39:20 +10:00
Daniel Han
0eb55f4290 Fix GGUF (#654)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

* Fix breaking bug in save.py with interpreting quantization_method as a string when saving to gguf (#651)

* Nightly (#649)

* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Fix bug in save.py with interpreting quantization_method as a string that prevents GGUF from saving

* Implemented better list management and then forgot to actually call the new list variable, fixed

* Check type of given quantization method and return type error if not list or string

* Update save.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>

* Revert "Fix breaking bug in save.py with interpreting quantization_method as …" (#652)

This reverts commit 506cb68867296237e95bc53c32f1bfc9b1757960.

* Revert "Revert "Fix breaking bug in save.py with interpreting quantization_me…" (#653)

This reverts commit 2f48cc9af385579876fd45bd833169d1f1a2ea58.

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
Co-authored-by: ArcadaLabs-Jason <52756218+ArcadaLabs-Jason@users.noreply.github.com>
2024-06-16 14:51:58 +10:00
Daniel Han
59fffc57de Nightly (#649)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

* check PEFT and base

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update chat_templates.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
2024-06-16 04:32:21 +10:00
Daniel Han
6a673412ec Nightly (#648)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

* Update gemma.py

* gpu

* Multiple GGUF saving

* Update save.py

* Update save.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
2024-06-16 03:39:00 +10:00
Daniel Han
b1cc523bda Nightly (#646)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

* Update llama.py

* Update llama.py

* GPU support

* Typo

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
2024-06-15 18:26:25 +10:00
Daniel Han
96e2fa423e Fix segfaults (#641)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Update mistral.py

* Auto check rope scaling

* Update llama.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
2024-06-15 00:52:33 +10:00
Daniel Han
9116eef815 Qwen bug fixes (#639)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

* Update qwen2.py

* docs

* Update README.md

* Update qwen2.py

* README: Fix minor typo. (#559)

* README: Fix minor typo.

One-character typo fix while reading.

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update mistral.py

* Update qwen2.py

* Update qwen2.py

* Update qwen2.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update README.md

* FastMistralModel

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
Co-authored-by: Walter Korman <lemurware@gmail.com>
2024-06-14 20:59:45 +10:00
Daniel Han-Chen
fffa36d6c9 Update __init__.py 2024-06-14 16:00:26 +10:00
Daniel Han-Chen
0dc7c1b033 Update tokenizer_utils.py 2024-06-14 03:24:28 +10:00
Daniel Han
83671ff13b Nightly (#632)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

* Update tokenizer_utils.py

* fix case where gguf saving fails due to first_conversion dtype (#630)

* Support revision parameter in FastLanguageModel.from_pretrained (#629)

* support `revision` parameter

* match unsloth formatting of named parameters

* clears any selected_adapters before calling internal_model.save_pretrained (#609)

* Update __init__.py (#602)

Check for incompatible modules before importing unsloth

* Fixed unsloth/tokenizer_utils.py for chat training (#604)

* Add GGML saving option to Unsloth for easier Ollama model creation and testing. (#345)

* Add save to llama.cpp GGML to save.py.

* Fix conversion command and path of convert to GGML function.

* Add autosaving lora to the GGML function

* Create lora save function for conversion to GGML

* Test fix #2 for saving lora

* Test fix #3 to save  the lora adapters to convert to GGML

* Remove unwated tokenizer saving for conversion to ggml and added a few print statements.

* Needed tokenizer for saving, added it back, also made it more unslothy style by having positional arguments, and added a few messages.

* Positional arguments didn't work out, so reverted to older version of the code, and added a few comments.

* Test fix 1 for arch

* Test fix 2 new Mistral error.

* Test fix 3

* Revert to old version for testing.

* Upload issue test fix 1

* Fix 2 uploading ggml

* Positional ags added.

* Temporray remove positional args

* Fix upload again!!!

* Add print statements and fix link

* Make the calling name better

* Create local saving for GGML

* Add choosing directory to save local GGML.

* Fix lil variable error in the save_to_custom_dir func

* docs: Add LoraConfig parameters documentation (#619)

* llama.cpp failing (#371)

llama.cpp is failing to generate quantize versions for the trained models.

Error:

```bash
You might have to compile llama.cpp yourself, then run this again.
You do not need to close this Python program. Run the following commands in a new terminal:
You must run this in the same folder as you're saving your model.
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp && make clean && LLAMA_CUDA=1 make all -j
Once that's done, redo the quantization.
```

But when i do clone this with recursive it works.

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* fix libcuda_dirs import for triton 3.0 (#227)

* fix libcuda_dirs import for triton 3.0

* Update __init__.py

* Update __init__.py

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update __init__.py

* Update fast_lora.py

* Update save.py

* Update save.py

* Update save.py

* Update loader.py

* Update save.py

* Update save.py

* quantize now llama-quantize

* Update chat_templates.py

* Update loader.py

* Update mapper.py

* Update __init__.py

* embedding size

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Eliot Hall <60240707+chrehall68@users.noreply.github.com>
Co-authored-by: Rickard Edén <rickardeden@gmail.com>
Co-authored-by: XiaoYang <xyangk@gmail.com>
Co-authored-by: Oseltamivir <58582368+Oseltamivir@users.noreply.github.com>
Co-authored-by: mahiatlinux <110882203+mahiatlinux@users.noreply.github.com>
Co-authored-by: Sébastien De Greef <sebdg@binarycompany.com>
Co-authored-by: Alberto Ferrer <albertof@barrahome.org>
Co-authored-by: Thomas Viehmann <tv.github-private@beamnet.de>
2024-06-14 02:58:08 +10:00
Daniel Han
23a8f5809d Ollama Chat Templates (#582)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

* Update save.py

* Update save.py

* Update save.py

* Update save.py

* remove_special_tokens

* Ollama

* Update chat_templates.py

* Update chat_templates.py

* Update chat_templates.py

* Update llama.py

* Update chat_templates.py

* Support bfloat16 GGUF

* Update save.py

* Update llama.py

* fast_forward_inference

* Update mapper.py

* Update loader.py

* Update llama.py

* Update tokenizer_utils.py

* info

* edits

* Create chat template

* Fix tokenizer

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-06-13 05:04:54 +10:00
Daniel Han-Chen
4773950b65 Update llama.py 2024-06-07 04:54:57 +10:00
Daniel Han-Chen
82520f5e4c Update llama.py 2024-06-07 04:25:33 +10:00
Daniel Han-Chen
bc2b9f11d8 Update utils.py 2024-06-07 03:47:44 +10:00
Daniel Han-Chen
23e4103895 Qwen2 2024-06-07 02:53:17 +10:00
Daniel Han-Chen
9989591817 Update pyproject.toml 2024-06-06 01:22:29 +10:00
Daniel Han-Chen
d2fa16aafa Update llama.py 2024-06-05 20:57:08 +10:00
Daniel Han-Chen
0eab5152d2 Update README.md 2024-06-05 06:15:27 +10:00
Daniel Han-Chen
b52f372944 Update README.md 2024-06-05 06:14:11 +10:00
Daniel Han
1b798b3a97 Fix #563 (#564)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* accelerate

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update tokenizer_utils.py

* train_dataloader

* Update llama.py

* Update llama.py

* Update llama.py

* use_fast_convert

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-05-31 00:41:44 +10:00
Daniel Han
d86f18a2d6 Fix Phi-3 (#556)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

* Update _utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-05-29 14:30:26 +10:00
Daniel Han
4ed58ab212 Nightly (#548)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3

* Update save.py

* Update README.md

Mistral v3 to Mistral v0.3

* Untrained tokens

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update save.py

* Update save.py

* Update save.py

* checkpoint

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-05-29 00:30:38 +10:00
Daniel Han-Chen
13d5917415 Update tokenizer_utils.py 2024-05-24 11:21:29 +10:00
Z
a2180942b6 Update _utils.py (#520)
Fixed a typo in the tokenizer fixer.
2024-05-24 11:10:06 +10:00
Daniel Han
32b5f8daf0 Phi-3, Llama-3 bug fixes (#519)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py

* Phi-3
2024-05-24 06:36:10 +10:00
Daniel Han
57bb5ad0f6 Phi 3 Medium (#518)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3

* Phi 3 medium

* Update chat_templates.py

* Update chat_templates.py
2024-05-24 04:24:01 +10:00
Daniel Han-Chen
425655ecec Update README.md 2024-05-23 04:31:26 +10:00
Daniel Han
d0b92ea58b Mistral v3 (#514)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py

* Mistral v3
2024-05-23 04:15:02 +10:00
Daniel Han
3b71dd7300 Fix is_bfloat16_supported missing (#510)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py

* is_bfloat16_supported

* Update __init__.py
2024-05-22 20:40:57 +10:00
Daniel Han
5fc27f1f96 Nightly (#506)
* Update llama.py

* offload

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* continued pretraining trainer

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* is_bfloat16_supported

* Update __init__.py

* Update README.md

* Update llama.py
2024-05-22 04:45:57 +10:00
Daniel Han
0eb57384bd Nightly (#483)
* peft issue

* Update save.py

* Update __init__.py

* Update pyproject.toml
2024-05-17 23:46:33 +10:00
Daniel Han-Chen
0c2f3952b7 Squashed commit of the following:
commit 23bd794c246e9c90c453c9f2ab41a21ac1e41b9d
Author: Daniel Han-Chen <danielhanchen@gmail.com>
Date:   Fri May 17 14:16:39 2024 +1000

    Update save.py

commit c12cc6c2b13333c4c6709e0ee88665b08c672887
Author: Daniel Han-Chen <danielhanchen@gmail.com>
Date:   Fri May 17 04:14:05 2024 +1000

    peft issue
2024-05-17 14:18:52 +10:00
Daniel Han
774002fc46 peft issue (#480) 2024-05-17 04:18:18 +10:00
Daniel Han
c7e9158b29 Fix generation (#472)
* Fix prompt

* Update chat_templates.py

* fix_untrained_tokens

* Update llama.py

* add tokens

* Update _utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* pad_token

* Update chat_templates.py

* Update chat_templates.py

* tokenizer

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* patch tokenizer padding

* Update tokenizer_utils.py

* Update save.py

* Fix: loading models with resized vocabulary (#377)

* new: vocab resize on load

* new: gitignore

* GGUF fix

* Readme (#390)

* Update README.md

* Update README.md

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>

* Update README.md

* Delete .gitignore

* Phi-3

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Fix reserved tokens

* Update save.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update chat_templates.py

* Update save.py

* Update _utils.py

* Update chat_templates.py

* Adds dependencies and extras for torch 2.3.0 with new xformers versions (#415)

* Adds dependencies and extras for torch 2.3.0 with new xformers versions

* Add 2.3.0 section to readme

* Support Qwen2 (#428)

* support Qwen2

* support Qwen2

* Delete README.md

* Revert "Delete README.md"

This reverts commit 9dde82c35d446393946c3497ad5cf96a2b59197e.

* Update README.md

* Qwen2 == Mistral

* Update llama.py

* Update __init__.py

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update save.py

* Update _utils.py

* Update save.py

* Update save.py

* Update save.py

* test_hf_gguf_equivalence

* Update chat_templates.py

* Update chat_templates.py

* --pad-vocab

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Unspecified max_seq_length

* possible_pad_token

* Update tokenizer_utils.py

* past_key_values

* Update llama.py

* Update llama.py

* Update llama.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* Update _utils.py

* _wrap_fast_inference

* Update llama.py

* Update llama.py

* flag

---------

Co-authored-by: Igor Kilbas <whitemarsstudios@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nathan Azrak <42650258+nathan-az@users.noreply.github.com>
Co-authored-by: Yang JianXin <995462226@qq.com>
2024-05-16 15:09:42 +10:00
Daniel Han
b61caacece Nightly (#461)
* Fix prompt

* Update chat_templates.py

* fix_untrained_tokens

* Update llama.py

* add tokens

* Update _utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* pad_token

* Update chat_templates.py

* Update chat_templates.py

* tokenizer

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* patch tokenizer padding

* Update tokenizer_utils.py

* Update save.py

* Fix: loading models with resized vocabulary (#377)

* new: vocab resize on load

* new: gitignore

* GGUF fix

* Readme (#390)

* Update README.md

* Update README.md

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>

* Update README.md

* Delete .gitignore

* Phi-3

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Fix reserved tokens

* Update save.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update chat_templates.py

* Update save.py

* Update _utils.py

* Update chat_templates.py

* Adds dependencies and extras for torch 2.3.0 with new xformers versions (#415)

* Adds dependencies and extras for torch 2.3.0 with new xformers versions

* Add 2.3.0 section to readme

* Support Qwen2 (#428)

* support Qwen2

* support Qwen2

* Delete README.md

* Revert "Delete README.md"

This reverts commit 9dde82c35d446393946c3497ad5cf96a2b59197e.

* Update README.md

* Qwen2 == Mistral

* Update llama.py

* Update __init__.py

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update save.py

* Update _utils.py

* Update save.py

* Update save.py

* Update save.py

* test_hf_gguf_equivalence

* Update chat_templates.py

* Update chat_templates.py

* --pad-vocab

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Unspecified max_seq_length

* possible_pad_token

* Update tokenizer_utils.py

---------

Co-authored-by: Igor Kilbas <whitemarsstudios@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nathan Azrak <42650258+nathan-az@users.noreply.github.com>
Co-authored-by: Yang JianXin <995462226@qq.com>
2024-05-14 04:51:23 +10:00
Daniel Han
66fab81a26 May 2024 Prelim (#447)
* Fix prompt

* Update chat_templates.py

* fix_untrained_tokens

* Update llama.py

* add tokens

* Update _utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* pad_token

* Update chat_templates.py

* Update chat_templates.py

* tokenizer

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* patch tokenizer padding

* Update tokenizer_utils.py

* Update save.py

* Fix: loading models with resized vocabulary (#377)

* new: vocab resize on load

* new: gitignore

* GGUF fix

* Readme (#390)

* Update README.md

* Update README.md

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>

* Update README.md

* Delete .gitignore

* Phi-3

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Fix reserved tokens

* Update save.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update chat_templates.py

* Update save.py

* Update _utils.py

* Update chat_templates.py

* Adds dependencies and extras for torch 2.3.0 with new xformers versions (#415)

* Adds dependencies and extras for torch 2.3.0 with new xformers versions

* Add 2.3.0 section to readme

* Support Qwen2 (#428)

* support Qwen2

* support Qwen2

* Delete README.md

* Revert "Delete README.md"

This reverts commit 9dde82c35d446393946c3497ad5cf96a2b59197e.

* Update README.md

* Qwen2 == Mistral

* Update llama.py

* Update __init__.py

* Update README.md

---------

Co-authored-by: Daniel Han <danielhanchen@gmail.com>

* Update save.py

* Update save.py

* Update _utils.py

* Update save.py

* Update save.py

* Update save.py

* test_hf_gguf_equivalence

* Update chat_templates.py

* Update chat_templates.py

* --pad-vocab

* Update tokenizer_utils.py

---------

Co-authored-by: Igor Kilbas <whitemarsstudios@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nathan Azrak <42650258+nathan-az@users.noreply.github.com>
Co-authored-by: Yang JianXin <995462226@qq.com>
2024-05-13 05:22:03 +10:00
Daniel Han
ec171ad1d7 llama-3 bug fixes (#429)
* Fix prompt

* Update chat_templates.py

* fix_untrained_tokens

* Update llama.py

* add tokens

* Update _utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* pad_token

* Update chat_templates.py

* Update chat_templates.py

* tokenizer

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* patch tokenizer padding

* Update tokenizer_utils.py

* Update save.py

* Fix: loading models with resized vocabulary (#377)

* new: vocab resize on load

* new: gitignore

* GGUF fix

* Readme (#390)

* Update README.md

* Update README.md

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>

* Update README.md

* Delete .gitignore

* Phi-3

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Fix reserved tokens

* Update save.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update chat_templates.py

* Update save.py

* Update _utils.py

* Update chat_templates.py

---------

Co-authored-by: Igor Kilbas <whitemarsstudios@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-05-08 07:40:41 +10:00
Daniel Han-Chen
2ae753b719 Update save.py 2024-05-05 13:28:21 +10:00
Daniel Han
4a802a4bb1 Fix llama-3 (#423)
* Fix prompt

* Update chat_templates.py

* fix_untrained_tokens

* Update llama.py

* add tokens

* Update _utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* pad_token

* Update chat_templates.py

* Update chat_templates.py

* tokenizer

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* patch tokenizer padding

* Update tokenizer_utils.py

* Update save.py

* Fix: loading models with resized vocabulary (#377)

* new: vocab resize on load

* new: gitignore

* GGUF fix

* Readme (#390)

* Update README.md

* Update README.md

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>

* Update README.md

* Delete .gitignore

* Phi-3

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Fix reserved tokens

* Update save.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

* Update tokenizer_utils.py

---------

Co-authored-by: Igor Kilbas <whitemarsstudios@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-05-05 05:45:01 +10:00
Daniel Han-Chen
0560136e94 Update README.md 2024-04-30 20:26:10 +10:00
Daniel Han
6e4ebe4040 Phi-3 (#397)
* Fix prompt

* Update chat_templates.py

* fix_untrained_tokens

* Update llama.py

* add tokens

* Update _utils.py

* Update tokenizer_utils.py

* Update llama.py

* Update llama.py

* Update llama.py

* Update llama.py

* pad_token

* Update chat_templates.py

* Update chat_templates.py

* tokenizer

* Update save.py

* Update chat_templates.py

* Update chat_templates.py

* patch tokenizer padding

* Update tokenizer_utils.py

* Update save.py

* Fix: loading models with resized vocabulary (#377)

* new: vocab resize on load

* new: gitignore

* GGUF fix

* Readme (#390)

* Update README.md

* Update README.md

---------

Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>

* Update README.md

* Delete .gitignore

* Phi-3

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

* Update README.md

---------

Co-authored-by: Igor Kilbas <whitemarsstudios@gmail.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
2024-04-30 05:59:02 +10:00