Daniel Han
4feae9ae42
Update _utils.py
2025-01-08 15:48:40 -08:00
Daniel Han
1767be3692
Update tokenizer_utils.py
2025-01-08 15:48:11 -08:00
Daniel Han
3b4364985f
Phi-4 bug fix
2025-01-08 15:40:27 -08:00
Daniel Han
6cbfca8c63
Phi-4 ( #1523 )
...
* use exact model name
* Update save.py
* Update _utils.py
* Update _utils.py
* Update _utils.py
* Update _utils.py
* print
* Update _utils.py
* Update _utils.py
* Update llama.py
* Update _utils.py
* Update vision.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 loader.py
* accurate_accumulation
* Update loader.py
* Update loader.py
* Update _utils.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update pyproject.toml
* Update __init__.py
* Update pyproject.toml
* Update __init__.py
* Update __init__.py
* Fix Triton heuristics
https://github.com/triton-lang/triton/issues/5224
* Update __init__.py
* Update __init__.py
* Update __init__.py
* Update __init__.py
* Xformers
* Update loader.py
* Update loader.py
* Rewind
* Update _utils.py
* Update _utils.py
* requires grad
* Update loader.py
* Update _utils.py
* Update loader.py
* changing model to base_model if peft model is already used
* Improve debugging experience (#1512 )
* Create CONTRIBUTING.md (#1472 )
Creating contributing guidelines
* Update CONTRIBUTING.md
improved sentence
* Improve logging control in `unsloth_compile_transformers` by conditionally redirecting stdout based on UNSLOTH_DISABLE_LOGGER environment variable
---------
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nino Risteski <95188570+NinoRisteski@users.noreply.github.com>
* Update loader.py
* Update llama.py
* Update llama.py
* Revert "Update llama.py"
This reverts commit b7ddf962d2#1491 )
* change initilization of n_heads, n_kv_heads, hidden_size in llama.py
* do the same for cohere, mistral, gemma2, granite
* do the same for flexattention,cohere, mistral, granite
* Update llama.py
* Update llama.py
* Update granite to work with latest post_patch methods (#1502 )
* Update granite to work with latest post_patch methods
* Pass position_embeddings for granite even if transformers<4.47
* Update llama.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* Minor fixes for granite models (#1503 )
* Update granite.py
Grab residual multiplier directly from layer
* Update llama.py
Version should read >= 4.47.1 as that is the version requiring the changes
* Update granite.py
* Update llama.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* support modelscope models and datasets (#1481 )
* support modelscope
* change modelscope args
* remove useless import
* remove useless import
* fix
* wip
* fix
* remove useless code
* add readme
* add some comments
* change print to raise error
* update comment
* Update loader.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* Merge branch 'main' into nightly
* Phi 4
---------
Co-authored-by: Itsuro Tajima <tajima@georepublic.de>
Co-authored-by: Muhammad Osama <muhammadosama1994@gmail.com>
Co-authored-by: Edd <68678137+Erland366@users.noreply.github.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nino Risteski <95188570+NinoRisteski@users.noreply.github.com>
Co-authored-by: Kareem <81531392+KareemMusleh@users.noreply.github.com>
Co-authored-by: Datta Nimmaturi <datta.nimmaturi@nutanix.com>
Co-authored-by: Z <coffeevampirebusiness@gmail.com>
Co-authored-by: tastelikefeet <58414341+tastelikefeet@users.noreply.github.com>
2025-01-08 15:10:46 -08:00
Daniel Han
1820995bae
Merge branch 'main' into nightly
2025-01-08 15:10:13 -08:00
Daniel Han
f77d6d608c
Phi 4
2025-01-08 14:38:41 -08:00
Daniel Han
0554918864
Merge branch 'main' into nightly
2025-01-08 12:42:18 -08:00
sebaxakerhtc
71ca60c7f0
Update __init__.py ( #1520 )
...
* Update __init__.py
This PR is solving the (issue)[https://github.com/unslothai/unsloth/issues/1518 ] with some GPUs
* Update __init__.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
2025-01-07 14:51:17 -08:00
Daniel Han
d90aefea98
Update pyproject.toml
2025-01-07 04:29:09 -08:00
Daniel Han
63782ea3af
Bug fixes ( #1516 )
...
* use exact model name
* Update save.py
* Update _utils.py
* Update _utils.py
* Update _utils.py
* Update _utils.py
* print
* Update _utils.py
* Update _utils.py
* Update llama.py
* Update _utils.py
* Update vision.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 loader.py
* accurate_accumulation
* Update loader.py
* Update loader.py
* Update _utils.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update loader.py
* Update pyproject.toml
* Update __init__.py
* Update pyproject.toml
* Update __init__.py
* Update __init__.py
* Fix Triton heuristics
https://github.com/triton-lang/triton/issues/5224
* Update __init__.py
* Update __init__.py
* Update __init__.py
* Update __init__.py
* Xformers
* Update loader.py
* Update loader.py
* Rewind
* Update _utils.py
* Update _utils.py
* requires grad
* Update loader.py
* Update _utils.py
* Update loader.py
* changing model to base_model if peft model is already used
* Improve debugging experience (#1512 )
* Create CONTRIBUTING.md (#1472 )
Creating contributing guidelines
* Update CONTRIBUTING.md
improved sentence
* Improve logging control in `unsloth_compile_transformers` by conditionally redirecting stdout based on UNSLOTH_DISABLE_LOGGER environment variable
---------
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nino Risteski <95188570+NinoRisteski@users.noreply.github.com>
* Update loader.py
* Update llama.py
* Update llama.py
* Revert "Update llama.py"
This reverts commit b7ddf962d2#1491 )
* change initilization of n_heads, n_kv_heads, hidden_size in llama.py
* do the same for cohere, mistral, gemma2, granite
* do the same for flexattention,cohere, mistral, granite
* Update llama.py
* Update llama.py
* Update granite to work with latest post_patch methods (#1502 )
* Update granite to work with latest post_patch methods
* Pass position_embeddings for granite even if transformers<4.47
* Update llama.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* Minor fixes for granite models (#1503 )
* Update granite.py
Grab residual multiplier directly from layer
* Update llama.py
Version should read >= 4.47.1 as that is the version requiring the changes
* Update granite.py
* Update llama.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
* support modelscope models and datasets (#1481 )
* support modelscope
* change modelscope args
* remove useless import
* remove useless import
* fix
* wip
* fix
* remove useless code
* add readme
* add some comments
* change print to raise error
* update comment
* Update loader.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
---------
Co-authored-by: Itsuro Tajima <tajima@georepublic.de>
Co-authored-by: Muhammad Osama <muhammadosama1994@gmail.com>
Co-authored-by: Edd <68678137+Erland366@users.noreply.github.com>
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nino Risteski <95188570+NinoRisteski@users.noreply.github.com>
Co-authored-by: Kareem <81531392+KareemMusleh@users.noreply.github.com>
Co-authored-by: Datta Nimmaturi <datta.nimmaturi@nutanix.com>
Co-authored-by: Z <coffeevampirebusiness@gmail.com>
Co-authored-by: tastelikefeet <58414341+tastelikefeet@users.noreply.github.com>
2025-01-07 04:23:14 -08:00
tastelikefeet
83421fd2b5
support modelscope models and datasets ( #1481 )
...
* support modelscope
* change modelscope args
* remove useless import
* remove useless import
* fix
* wip
* fix
* remove useless code
* add readme
* add some comments
* change print to raise error
* update comment
* Update loader.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
2025-01-07 04:09:36 -08:00
Z
3cde4e1922
Minor fixes for granite models ( #1503 )
...
* Update granite.py
Grab residual multiplier directly from layer
* Update llama.py
Version should read >= 4.47.1 as that is the version requiring the changes
* Update granite.py
* Update llama.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
2025-01-07 03:58:40 -08:00
Datta Nimmaturi
c8e9dcf4f8
Update granite to work with latest post_patch methods ( #1502 )
...
* Update granite to work with latest post_patch methods
* Pass position_embeddings for granite even if transformers<4.47
* Update llama.py
---------
Co-authored-by: Daniel Han <danielhanchen@gmail.com>
2025-01-07 03:49:11 -08:00
Daniel Han
84fa1af77e
Update llama.py
2025-01-07 03:39:08 -08:00
Daniel Han
b7d47c1d8a
Update llama.py
2025-01-07 03:33:46 -08:00
Kareem
0e2110f7b8
All attention refactor fix ( #1491 )
...
* change initilization of n_heads, n_kv_heads, hidden_size in llama.py
* do the same for cohere, mistral, gemma2, granite
* do the same for flexattention,cohere, mistral, granite
2025-01-07 02:41:15 -08:00
Michael Han
4ce92cfe2c
Update README.md
...
Notebook links
2025-01-07 02:02:59 -08:00
Daniel Han
a4aba47ebb
Update llama.py
2025-01-07 01:56:32 -08:00
Daniel Han
0672e71b17
Force data-type
2025-01-07 01:51:49 -08:00
Daniel Han
6320381fb6
Update llama.py
2025-01-07 01:43:20 -08:00
Daniel Han
ab2b72c5f0
Auto change is_bfloat16_supported
2025-01-07 01:40:04 -08:00
Daniel Han
9e00262be6
Update llama.py
2025-01-07 01:10:41 -08:00
Daniel Han
4df3af2f57
Update llama.py
2025-01-07 01:05:04 -08:00
Daniel Han
358316522f
Update llama.py
2025-01-07 01:02:35 -08:00
Daniel Han
97c3e282fb
Update llama.py
2025-01-07 01:02:24 -08:00
Daniel Han
020c793a1e
Update llama.py
2025-01-07 00:50:56 -08:00
Daniel Han
656099cfc3
Update llama.py
2025-01-07 00:45:22 -08:00
Daniel Han
837d620dbd
Update llama.py
2025-01-07 00:41:26 -08:00
Daniel Han
c95380cc9c
Update llama.py
2025-01-07 00:41:16 -08:00
Daniel Han
689ca57214
Update llama.py
2025-01-07 00:38:04 -08:00
Daniel Han
dc33cc94a7
Update llama.py
2025-01-07 00:34:46 -08:00
Daniel Han
f791766ab9
Update llama.py
2025-01-07 00:34:09 -08:00
Daniel Han
a7740ba8e9
Update llama.py
2025-01-07 00:30:32 -08:00
Daniel Han
883c25d34c
Merge branch 'pr/1509' into nightly
2025-01-06 22:08:07 -08:00
Daniel Han
c4720f1baf
Update llama.py
2025-01-06 22:06:00 -08:00
Daniel Han
294cd8ea32
Revert "Update llama.py"
...
This reverts commit a8edd0931a
2025-01-06 22:05:44 -08:00
Daniel Han
a8edd0931a
Update llama.py
2025-01-06 22:05:14 -08:00
Daniel Han
0f6b518ee1
Update llama.py
2025-01-06 18:56:26 -08:00
Daniel Han
adb2dcfd2b
Update loader.py
2025-01-06 18:13:48 -08:00
Edd
9940583287
Improve debugging experience ( #1512 )
...
* Create CONTRIBUTING.md (#1472 )
Creating contributing guidelines
* Update CONTRIBUTING.md
improved sentence
* Improve logging control in `unsloth_compile_transformers` by conditionally redirecting stdout based on UNSLOTH_DISABLE_LOGGER environment variable
---------
Co-authored-by: Michael Han <107991372+shimmyshimmer@users.noreply.github.com>
Co-authored-by: Nino Risteski <95188570+NinoRisteski@users.noreply.github.com>
2025-01-06 18:04:27 -08:00
Daniel Han
8cf3e6fa2b
Merge branch 'main' into nightly
2025-01-06 18:03:53 -08:00
Muhammad Osama
6b9e11bdf3
changing model to base_model if peft model is already used
2025-01-05 18:18:42 -06:00
Michael Han
48627f876c
Merge pull request #1507 from NinoRisteski/patch-1
...
Update CONTRIBUTING.md
2025-01-05 01:56:40 -08:00
Nino Risteski
8063abc004
Update CONTRIBUTING.md
...
improved sentence
2025-01-05 09:24:10 +01:00
Michael Han
fb49390494
Create CONTRIBUTING.md ( #1472 )
...
Creating contributing guidelines
2025-01-04 22:09:25 -08:00
Daniel Han
d08c8afd6c
Merge branch 'pr/1339' into nightly
2025-01-04 22:08:00 -08:00
Daniel Han
3e1c5ec3a0
Update loader.py
2025-01-04 22:03:11 -08:00
Daniel Han
c697d6d01a
Update _utils.py
2025-01-04 00:42:39 -08:00
Daniel Han
5cf47b3e63
Update loader.py
2025-01-02 22:44:17 -08:00
Daniel Han
75ffad921f
requires grad
2025-01-02 18:25:25 -08:00
