Merge branch 'main' into nightly
This commit is contained in:
Daniel Han
commit
ddd4e86cec
5 changed files with 20 additions and 13 deletions
10
README.md
10
README.md
|
|
@ -6,11 +6,11 @@
|
|||
<img alt="unsloth logo" src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20logo%20black%20text.png" height="110" style="max-width: 100%;">
|
||||
</picture></a>
|
||||
|
||||
<a href="https://colab.research.google.com/drive/135ced7oHytdxu3N2DNe1Z0kqjyYIkDXp?usp=sharing"><img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/start free finetune button.png" height="48"></a>
|
||||
<a href="https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing"><img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/start free finetune button.png" height="48"></a>
|
||||
<a href="https://discord.gg/unsloth"><img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/Discord button.png" height="48"></a>
|
||||
<a href="https://ko-fi.com/unsloth"><img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/buy me a coffee button.png" height="48"></a>
|
||||
|
||||
### Finetune Llama 3, Mistral, Phi-3 & Gemma 2-5x faster with 80% less memory!
|
||||
### Finetune Llama 3.1, Mistral, Phi-3 & Gemma 2-5x faster with 80% less memory!
|
||||
|
||||
|
||||
|
||||
|
|
@ -22,7 +22,7 @@ All notebooks are **beginner friendly**! Add your dataset, click "Run All", and
|
|||
|
||||
| Unsloth supports | Free Notebooks | Performance | Memory use |
|
||||
|-----------|---------|--------|----------|
|
||||
| **Llama 3.1 (8B)** | [▶️ Start for free](https://colab.research.google.com/drive/135ced7oHytdxu3N2DNe1Z0kqjyYIkDXp?usp=sharing) | 2x faster | 60% less |
|
||||
| **Llama 3.1 (8B)** | [▶️ Start for free](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) | 2x faster | 60% less |
|
||||
| **Mistral Nemo (12B)** | [▶️ Start for free](https://colab.research.google.com/drive/17d3U-CAIwzmbDRqbZ9NnpHxCkmXB6LZ0?usp=sharing) | 2x faster | 60% less |
|
||||
| **Gemma 2 (9B)** | [▶️ Start for free](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) | 2x faster | 63% less |
|
||||
| **Phi-3 (mini)** | [▶️ Start for free](https://colab.research.google.com/drive/1lN6hPQveB_mHSnTOYifygFcrO8C1bxq4?usp=sharing) | 2x faster | 50% less |
|
||||
|
|
@ -32,14 +32,14 @@ All notebooks are **beginner friendly**! Add your dataset, click "Run All", and
|
|||
| **DPO Zephyr** | [▶️ Start for free](https://colab.research.google.com/drive/15vttTpzzVXv_tJwEk-hIcQ0S9FcEWvwP?usp=sharing) | 1.9x faster | 43% less |
|
||||
| **TinyLlama** | [▶️ Start for free](https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing) | 3.9x faster | 74% less |
|
||||
|
||||
- **Kaggle Notebooks** for [Llama 3.1 (8B)](https://www.kaggle.com/code/danielhanchen/kaggle-llama-3-8b-unsloth-notebook), [Gemma 2 (9B)](https://www.kaggle.com/code/danielhanchen/kaggle-gemma-7b-unsloth-notebook/), [Mistral (7B)](https://www.kaggle.com/code/danielhanchen/kaggle-mistral-7b-unsloth-notebook)
|
||||
- **Kaggle Notebooks** for [Llama 3.1 (8B)](https://www.kaggle.com/danielhanchen/kaggle-llama-3-1-8b-unsloth-notebook), [Gemma 2 (9B)](https://www.kaggle.com/code/danielhanchen/kaggle-gemma-7b-unsloth-notebook/), [Mistral (7B)](https://www.kaggle.com/code/danielhanchen/kaggle-mistral-7b-unsloth-notebook)
|
||||
- Run [Llama 3 conversational notebook](https://colab.research.google.com/drive/1XamvWYinY6FOSX9GLvnqSjjsNflxdhNc?usp=sharing) and [Mistral v0.3 ChatML](https://colab.research.google.com/drive/15F1xyn8497_dUbxZP4zWmPZ3PJx1Oymv?usp=sharing)
|
||||
- This [text completion notebook](https://colab.research.google.com/drive/1ef-tab5bhkvWmBOObepl1WgJvfvSzn5Q?usp=sharing) is for continued pretraining / raw text
|
||||
- This [continued pretraining notebook](https://colab.research.google.com/drive/1tEd1FrOXWMnCU9UIvdYhs61tkxdMuKZu?usp=sharing) is for learning another language
|
||||
- Click [here](https://github.com/unslothai/unsloth/wiki) for detailed documentation for Unsloth.
|
||||
|
||||
## 🦥 Unsloth.ai News
|
||||
- 📣 NEW! [Llama 3.1 8b, 70b](https://colab.research.google.com/drive/135ced7oHytdxu3N2DNe1Z0kqjyYIkDXp?usp=sharing) both Base and Instruct now supported
|
||||
- 📣 NEW! [Llama 3.1 8b, 70b](https://colab.research.google.com/drive/1Ys44kVvmeZtnICzWz0xgpRnrIOjZAuxp?usp=sharing) both Base and Instruct now supported
|
||||
- 📣 NEW! [Mistral Nemo-12b](https://colab.research.google.com/drive/17d3U-CAIwzmbDRqbZ9NnpHxCkmXB6LZ0?usp=sharing) both Base and Instruct now supported
|
||||
- 📣 NEW! [Gemma-2-9b](https://colab.research.google.com/drive/1vIrqH5uYDQwsJ4-OO3DErvuv4pBgVwk4?usp=sharing) and Gemma-2-27b now supported
|
||||
- 📣 UPDATE! [Phi-3 mini](https://colab.research.google.com/drive/1hhdhBa1j_hsymiW9m-WzxQtgqTH_NHqi?usp=sharing) model updated. [Phi-3 Medium](https://colab.research.google.com/drive/1hhdhBa1j_hsymiW9m-WzxQtgqTH_NHqi?usp=sharing) 2x faster finetuning.
|
||||
|
|
|
|||
BIN
images/Run.png
Normal file
BIN
images/Run.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 11 KiB |
|
|
@ -393,7 +393,10 @@ def patch_tokenizer(model, tokenizer):
|
|||
tokenizer.pad_token = possible_pad_token
|
||||
if model is not None:
|
||||
config = model.config.update({"pad_token_id" : tokenizer.pad_token_id})
|
||||
pass
|
||||
else:
|
||||
if model is not None:
|
||||
if model.config.pad_token_id is None:
|
||||
config = model.config.update({"pad_token_id" : tokenizer.pad_token_id})
|
||||
return model, tokenizer
|
||||
pass
|
||||
|
||||
|
|
|
|||
|
|
@ -1178,10 +1178,11 @@ def _wrap_fast_inference(generate, device_type, dtype, model):
|
|||
kwargs.pop("token_type_ids", None)
|
||||
|
||||
# Check pad_token
|
||||
kwargs["pad_token_id"] = kwargs.pop(
|
||||
"pad_token_id",
|
||||
getattr(model.config, "eos_token_id", None),
|
||||
)
|
||||
model_eos_token_id = getattr(model.config, "eos_token_id", None)
|
||||
if model_eos_token_id is not None and hasattr(model_eos_token_id, "__iter__"):
|
||||
model_eos_token_id = model_eos_token_id[0]
|
||||
|
||||
kwargs["pad_token_id"] = kwargs.pop("pad_token_id", model_eos_token_id)
|
||||
|
||||
# Set pad token
|
||||
# old_pad_token_id = getattr(model.config, "pad_token_id", None)
|
||||
|
|
|
|||
|
|
@ -132,9 +132,12 @@ class FastLanguageModel(FastLlamaModel):
|
|||
model_type = model_config.model_type
|
||||
|
||||
if model_type == "llama":
|
||||
scaling_type1 = model_config.rope_scaling.get("type", None)
|
||||
scaling_type2 = model_config.rope_scaling.get("rope_type", None)
|
||||
scaling_type = scaling_type1 if scaling_type1 is not None else scaling_type2
|
||||
scaling_type = None
|
||||
if getattr(model_config, "rope_scaling", None) is not None:
|
||||
scaling_type1 = model_config.rope_scaling.get("type", None)
|
||||
scaling_type2 = model_config.rope_scaling.get("rope_type", None)
|
||||
scaling_type = scaling_type1 if scaling_type1 is not None else scaling_type2
|
||||
pass
|
||||
|
||||
if scaling_type == "llama3" and not SUPPORTS_LLAMA31:
|
||||
raise ImportError(
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue