# /// script # requires-python = ">=3.10,<3.14" # dependencies = [ # "accelerate", # "bitsandbytes>=0.43.0", # "datasets==4.3.0", # "hf_transfer", # "huggingface_hub>=0.34.0", # "marimo", # "peft", # "protobuf", # "sentencepiece", # "torchao>=0.16.0", # "transformers==4.56.2", # "triton>=3.2.0", # "trl==0.22.2", # "unsloth @ git+https://github.com/unslothai/unsloth", # "unsloth_zoo @ git+https://github.com/unslothai/unsloth-zoo", # ] # # [tool.uv] # no-build-package = [ # "bitsandbytes", # "triton", # "vllm", # "xformers", # ] # /// import marimo __generated_with = "0.23.8" app = marimo.App() @app.cell def _(): import marimo as mo return (mo,) @app.cell(hide_code=True) def _(mo): mo.md(r""" To run this notebook on your A100 molab Pro instance, hit the **▶ Run all** button in the bottom-right corner - or use `Ctrl/Cmd + Shift + R`.

Join Discord if you need help + ⭐ Star us on Github ⭐

To install Unsloth on your local device, follow [our guide](https://unsloth.ai/docs/get-started/install). This notebook is licensed [LGPL-3.0](https://github.com/unslothai/notebooks?tab=LGPL-3.0-1-ov-file#readme). You will learn how to do [data prep](#Data), how to [train](#Train), how to [run the model](#Inference), & how to save it """) return @app.cell(hide_code=True) def _(mo): mo.md(r""" ### News """) return @app.cell(hide_code=True) def _(mo): mo.md(r""" Introducing **Unsloth Studio** - a new open source, no-code web UI to train and run LLMs. [Blog](https://unsloth.ai/docs/new/studio) • [Notebook](https://github.com/unslothai/unsloth/blob/main/studio/Unsloth_Studio_Colab.ipynb)

_{Train models — no code needed}

_{Run GGUF models on Mac, Windows & Linux}

Train MoEs - DeepSeek, GLM, Qwen and gpt-oss 12x faster with 35% less VRAM. [Blog](https://unsloth.ai/docs/new/faster-moe) Ultra Long-Context Reinforcement Learning is here with 7x more context windows! [Blog](https://unsloth.ai/docs/new/grpo-long-context) New in Reinforcement Learning: [FP8 RL](https://unsloth.ai/docs/new/fp8-reinforcement-learning) • [Vision RL](https://unsloth.ai/docs/new/vision-reinforcement-learning-vlm-rl) • [Standby](https://unsloth.ai/docs/basics/memory-efficient-rl) • [gpt-oss RL](https://unsloth.ai/docs/new/gpt-oss-reinforcement-learning) Visit our docs for all our [model uploads](https://unsloth.ai/docs/get-started/unsloth-model-catalog) and [notebooks](https://unsloth.ai/docs/get-started/unsloth-notebooks). """) return @app.cell(hide_code=True) def _(mo): mo.md(r""" ### Unsloth """) return @app.cell def _(): from unsloth import FastLanguageModel import torch max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally! dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+ load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False. # 4bit pre quantized models we support for 4x faster downloading + no OOMs. fourbit_models = [ "unsloth/Llama-3.1-8B-bnb-4bit", # Llama-3.1 2x faster "unsloth/Llama-3.1-8B-Instruct-bnb-4bit", "unsloth/Llama-3.1-70B-bnb-4bit", "unsloth/Llama-3.1-405B-bnb-4bit", # 4bit for 405b! "unsloth/Mistral-Small-Instruct-2409", # Mistral 22b 2x faster! "unsloth/mistral-7b-instruct-v0.3-bnb-4bit", "unsloth/Phi-3.5-mini-instruct", # Phi-3.5 2x faster! "unsloth/Phi-3-medium-4k-instruct", "unsloth/gemma-2-9b-bnb-4bit", "unsloth/gemma-2-27b-bnb-4bit", # Gemma 2x faster! "unsloth/Llama-3.2-1B-bnb-4bit", # NEW! Llama 3.2 models "unsloth/Llama-3.2-1B-Instruct-bnb-4bit", "unsloth/Llama-3.2-3B-bnb-4bit", "unsloth/Llama-3.2-3B-Instruct-bnb-4bit", "unsloth/Llama-3.3-70B-Instruct-bnb-4bit", # NEW! Llama 3.3 70B! ] # More models at https://huggingface.co/unsloth model, tokenizer = FastLanguageModel.from_pretrained( model_name="unsloth/Llama-3.3-70B-Instruct", # or choose "unsloth/Llama-3.2-1B-Instruct" max_seq_length=max_seq_length, # Choose any! We auto support RoPE Scaling internally! dtype=dtype, # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+ load_in_4bit=load_in_4bit, # Use 4bit quantization to reduce memory usage. Can be False. # token = "YOUR_HF_TOKEN", # HF Token for gated models ) return ( FastLanguageModel, dtype, load_in_4bit, max_seq_length, model, tokenizer, torch, ) @app.cell(hide_code=True) def _(mo): mo.md(r""" We now add LoRA adapters so we only need to update 1 to 10% of all parameters! """) return @app.cell def _(FastLanguageModel, model): model_1 = FastLanguageModel.get_peft_model( model, r=16, # Choose any number > 0 ! Suggested 8, 16, 32, 64, 128 target_modules=[ "q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj", ], lora_alpha=16, lora_dropout=0, # Supports any, but = 0 is optimized bias="none", # Supports any, but = "none" is optimized use_gradient_checkpointing="unsloth", # True or "unsloth" for very long context random_state=3407, use_rslora=False, # We support rank stabilized LoRA loftq_config=None, # And LoftQ ) return (model_1,) @app.cell(hide_code=True) def _(mo): mo.md(r""" ### Data Prep We now use the `Llama-3.1` format for conversation style finetunes. We use [Maxime Labonne's FineTome-100k](https://huggingface.co/datasets/mlabonne/FineTome-100k) dataset in ShareGPT style. But we convert it to HuggingFace's normal multiturn format `("role", "content")` instead of `("from", "value")`/ Llama-3 renders multi turn conversations like below: ``` <|begin_of_text|><|start_header_id|>user<|end_header_id|> Hello!<|eot_id|><|start_header_id|>assistant<|end_header_id|> Hey there! How are you?<|eot_id|><|start_header_id|>user<|end_header_id|> I'm great thanks!<|eot_id|> ``` We use our `get_chat_template` function to get the correct chat template. We support `zephyr, chatml, mistral, llama, alpaca, vicuna, vicuna_old, phi3, llama3` and more. """) return @app.cell def _(tokenizer): from unsloth.chat_templates import get_chat_template tokenizer_1 = get_chat_template(tokenizer, chat_template="llama-3.1") def formatting_prompts_func(examples): convos = examples["conversations"] texts = [ tokenizer_1.apply_chat_template( convo, tokenize=False, add_generation_prompt=False ) for convo in convos ] return {"text": texts} from datasets import load_dataset dataset = load_dataset("mlabonne/FineTome-100k", split="train") return dataset, formatting_prompts_func, get_chat_template, tokenizer_1 @app.cell(hide_code=True) def _(mo): mo.md(r""" We now use `standardize_sharegpt` to convert ShareGPT style datasets into HuggingFace's generic format. This changes the dataset from looking like: ``` {"from": "system", "value": "You are an assistant"} {"from": "human", "value": "What is 2+2?"} {"from": "gpt", "value": "It's 4."} ``` to ``` {"role": "system", "content": "You are an assistant"} {"role": "user", "content": "What is 2+2?"} {"role": "assistant", "content": "It's 4."} ``` """) return @app.cell def _(dataset, formatting_prompts_func): from unsloth.chat_templates import standardize_sharegpt dataset_1 = standardize_sharegpt(dataset) dataset_1 = dataset_1.map(formatting_prompts_func, batched=True) return (dataset_1,) @app.cell(hide_code=True) def _(mo): mo.md(r""" We look at how the conversations are structured for item 5: """) return @app.cell def _(dataset_1): dataset_1[5]["conversations"] return @app.cell(hide_code=True) def _(mo): mo.md(r""" And we see how the chat template transformed these conversations. **[Notice]** Llama 3.1 Instruct's default chat template default adds `"Cutting Knowledge Date: December 2023\nToday Date: 26 July 2024"`, so do not be alarmed! """) return @app.cell def _(dataset_1): dataset_1[5]["text"] return @app.cell(hide_code=True) def _(mo): mo.md(r""" ### Train the model Now let's train our model. We do 60 steps to speed things up, but you can set `num_train_epochs=1` for a full run, and turn off `max_steps=None`. We also support `DPOTrainer` and `GRPOTrainer` for reinforcement learning!! """) return @app.cell def _(dataset_1, max_seq_length, model_1, tokenizer_1): from trl import SFTConfig, SFTTrainer from transformers import DataCollatorForSeq2Seq trainer = SFTTrainer( model=model_1, tokenizer=tokenizer_1, train_dataset=dataset_1, dataset_text_field="text", max_seq_length=max_seq_length, # Choose any! We auto support RoPE Scaling internally! data_collator=DataCollatorForSeq2Seq(tokenizer=tokenizer_1), packing=False, # Can make training 5x faster for short sequences. args=SFTConfig( per_device_train_batch_size=2, gradient_accumulation_steps=4, warmup_steps=5, max_steps=60, learning_rate=0.0002, logging_steps=1, optim="adamw_8bit", weight_decay=0.001, lr_scheduler_type="linear", seed=3407, output_dir="outputs", report_to="none", # Use TrackIO/WandB etc ), ) return (trainer,) @app.cell(hide_code=True) def _(mo): mo.md(r""" We also use Unsloth's `train_on_completions` method to only train on the assistant outputs and ignore the loss on the user's inputs. """) return @app.cell def _(trainer): from unsloth.chat_templates import train_on_responses_only trainer_1 = train_on_responses_only( trainer, instruction_part="<|start_header_id|>user<|end_header_id|>\n\n", response_part="<|start_header_id|>assistant<|end_header_id|>\n\n", ) return (trainer_1,) @app.cell(hide_code=True) def _(mo): mo.md(r""" We verify masking is actually done: """) return @app.cell def _(tokenizer_1, trainer_1): tokenizer_1.decode(trainer_1.train_dataset[5]["input_ids"]) return @app.cell def _(tokenizer_1, trainer_1): space = tokenizer_1(" ", add_special_tokens=False).input_ids[0] tokenizer_1.decode( [space if x == -100 else x for x in trainer_1.train_dataset[5]["labels"]] ) return @app.cell(hide_code=True) def _(mo): mo.md(r""" We can see the System and Instruction prompts are successfully masked! """) return @app.cell def _(torch): # @title Show current memory stats gpu_stats = torch.cuda.get_device_properties(0) start_gpu_memory = round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) max_memory = round(gpu_stats.total_memory / 1024 / 1024 / 1024, 3) print(f"GPU = {gpu_stats.name}. Max memory = {max_memory} GB.") print(f"{start_gpu_memory} GB of memory reserved.") return max_memory, start_gpu_memory @app.cell def _(trainer_1): trainer_stats = trainer_1.train() return (trainer_stats,) @app.cell def _(max_memory, start_gpu_memory, torch, trainer_stats): # @title Show final memory and time stats used_memory = round(torch.cuda.max_memory_reserved() / 1024 / 1024 / 1024, 3) used_memory_for_lora = round(used_memory - start_gpu_memory, 3) used_percentage = round(used_memory / max_memory * 100, 3) lora_percentage = round(used_memory_for_lora / max_memory * 100, 3) print(f"{trainer_stats.metrics['train_runtime']} seconds used for training.") print( f"{round(trainer_stats.metrics['train_runtime'] / 60, 2)} minutes used for training." ) print(f"Peak reserved memory = {used_memory} GB.") print(f"Peak reserved memory for training = {used_memory_for_lora} GB.") print(f"Peak reserved memory % of max memory = {used_percentage} %.") print(f"Peak reserved memory for training % of max memory = {lora_percentage} %.") return @app.cell(hide_code=True) def _(mo): mo.md(r""" ### Inference Let's run the model! You can change the instruction and input - leave the output blank! We use `min_p = 0.1` and `temperature = 1.5`. Read this [Tweet](https://x.com/menhguin/status/1826132708508213629) for more information on why. """) return @app.cell def _(FastLanguageModel, get_chat_template, model_1, tokenizer_1): tokenizer_2 = get_chat_template(tokenizer_1, chat_template="llama-3.1") FastLanguageModel.for_inference(model_1) messages = [ { "role": "user", "content": "Continue the fibonacci sequence: 1, 1, 2, 3, 5, 8,", } ] inputs = tokenizer_2.apply_chat_template( messages, tokenize=True, add_generation_prompt=True, return_tensors="pt" ).to("cuda") outputs = model_1.generate( input_ids=inputs, max_new_tokens=64, use_cache=True, temperature=1.5, min_p=0.1 ) tokenizer_2.batch_decode( outputs ) return (tokenizer_2,) @app.cell(hide_code=True) def _(mo): mo.md(r""" You can also use a `TextStreamer` for continuous inference - so you can see the generation token by token, instead of waiting the whole time! """) return @app.cell def _(FastLanguageModel, model_1, tokenizer_2): FastLanguageModel.for_inference(model_1) # Enable native 2x faster inference messages_1 = [ { "role": "user", "content": "Continue the fibonacci sequence: 1, 1, 2, 3, 5, 8,", } ] inputs_1 = tokenizer_2.apply_chat_template( messages_1, tokenize=True, add_generation_prompt=True, return_tensors="pt" ).to("cuda") from transformers import TextStreamer text_streamer = TextStreamer(tokenizer_2, skip_prompt=True) _ = model_1.generate( input_ids=inputs_1, streamer=text_streamer, max_new_tokens=128, use_cache=True, temperature=1.5, min_p=0.1, ) # Must add for generation return (TextStreamer,) @app.cell(hide_code=True) def _(mo): mo.md(r""" ### Saving, loading finetuned models To save the final model as LoRA adapters, either use Hugging Face's `push_to_hub` for an online save or `save_pretrained` for a local save. **[NOTE]** This ONLY saves the LoRA adapters, and not the full model. To save to 16bit or GGUF, scroll down! """) return @app.cell def _(model_1, tokenizer_2): model_1.save_pretrained("llama_lora") tokenizer_2.save_pretrained("llama_lora") return @app.cell(hide_code=True) def _(mo): mo.md(r""" Now if you want to load the LoRA adapters we just saved for inference, set `False` to `True`: """) return @app.cell def _(TextStreamer, dtype, load_in_4bit, max_seq_length, model_1, tokenizer_2): if False: from unsloth import FastLanguageModel as _FastLanguageModel _model, _tokenizer = _FastLanguageModel.from_pretrained( model_name="llama_lora", # or choose "unsloth/Llama-3.2-1B-Instruct" max_seq_length=max_seq_length, # Choose any! We auto support RoPE Scaling internally! dtype=dtype, # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+ load_in_4bit=load_in_4bit, # Use 4bit quantization to reduce memory usage. Can be False. ) _FastLanguageModel.for_inference(_model) messages_2 = [ {"role": "user", "content": "Describe a tall tower in the capital of France."} ] inputs_2 = tokenizer_2.apply_chat_template( messages_2, tokenize=True, add_generation_prompt=True, return_tensors="pt" ).to("cuda") text_streamer_1 = TextStreamer(tokenizer_2, skip_prompt=True) _ = model_1.generate( input_ids=inputs_2, streamer=text_streamer_1, max_new_tokens=128, use_cache=True, temperature=1.5, min_p=0.1, ) # Must add for generation return @app.cell(hide_code=True) def _(mo): mo.md(r""" You can also use Hugging Face's `AutoPeftModelForCausalLM`. Only use this if you do not have `unsloth` installed. It can be hopelessly slow, since `4bit` model downloading is not supported, and Unsloth's **inference is 2x faster**. """) return @app.cell def _(load_in_4bit): if False: from peft import AutoPeftModelForCausalLM as _AutoPeftModelForCausalLM from transformers import AutoTokenizer as _AutoTokenizer _model = _AutoPeftModelForCausalLM.from_pretrained( "llama_lora", load_in_4bit=load_in_4bit ) _tokenizer = _AutoTokenizer.from_pretrained("llama_lora") return @app.cell(hide_code=True) def _(mo): mo.md(r""" ### Saving to float16 for VLLM We also support saving to `float16` directly. Select `merged_16bit` for float16 or `merged_4bit` for int4. We also allow `lora` adapters as a fallback. Use `push_to_hub_merged` to upload to your Hugging Face account! You can go to https://huggingface.co/settings/tokens for your personal tokens. See [our docs](https://unsloth.ai/docs/basics/inference-and-deployment) for more deployment options. """) return @app.cell def _(model_1, tokenizer_2): # Merge to 16bit if False: model_1.save_pretrained_merged( "llama_finetune_16bit", tokenizer_2, save_method="merged_16bit" ) if False: # Merge to 4bit model_1.push_to_hub_merged( "HF_USERNAME/llama_finetune_16bit", tokenizer_2, save_method="merged_16bit", token="YOUR_HF_TOKEN", # Get a token at https://huggingface.co/settings/tokens ) if False: model_1.save_pretrained_merged( "llama_finetune_4bit", tokenizer_2, save_method="merged_4bit" ) # Just LoRA adapters if False: model_1.push_to_hub_merged( "HF_USERNAME/llama_finetune_4bit", tokenizer_2, save_method="merged_4bit", token="YOUR_HF_TOKEN", # Get a token at https://huggingface.co/settings/tokens ) if False: model_1.save_pretrained("llama_lora") tokenizer_2.save_pretrained("llama_lora") if False: model_1.push_to_hub("HF_USERNAME/llama_lora", token="YOUR_HF_TOKEN") tokenizer_2.push_to_hub("HF_USERNAME/llama_lora", token="YOUR_HF_TOKEN") return @app.cell(hide_code=True) def _(mo): mo.md(r""" ### GGUF / llama.cpp Conversion To save to `GGUF` / `llama.cpp`, we support it natively now! We clone `llama.cpp` and we default save it to `q8_0`. We allow all methods like `q4_k_m`. Use `save_pretrained_gguf` for local saving and `push_to_hub_gguf` for uploading to HF. Some supported quant methods (full list on our [docs page](https://unsloth.ai/docs/basics/inference-and-deployment/saving-to-gguf)): * `q8_0` - Fast conversion. High resource use, but generally acceptable. * `q4_k_m` - Recommended. Uses Q6_K for half of the attention.wv and feed_forward.w2 tensors, else Q4_K. * `q5_k_m` - Recommended. Uses Q6_K for half of the attention.wv and feed_forward.w2 tensors, else Q5_K. [**NEW**] To finetune and auto export to Ollama, try our [Ollama notebook](https://github.com/unslothai/notebooks/blob/main/nb/Llama3_(8B)-Ollama.ipynb) """) return @app.cell def _(model_1, tokenizer_2): # Save to 8bit Q8_0 if False: # Remember to go to https://huggingface.co/settings/tokens for a token! # And change hf to your username! model_1.save_pretrained_gguf("llama_finetune", tokenizer_2) if False: # Save to 16bit GGUF model_1.push_to_hub_gguf( "HF_USERNAME/llama_finetune", tokenizer_2, token="YOUR_HF_TOKEN" ) if False: model_1.save_pretrained_gguf( "llama_finetune", tokenizer_2, quantization_method="f16" ) # Save to q4_k_m GGUF if False: model_1.push_to_hub_gguf( "HF_USERNAME/llama_finetune", tokenizer_2, quantization_method="f16", token="YOUR_HF_TOKEN", # Get a token at https://huggingface.co/settings/tokens ) if False: # Save to multiple GGUF options - much faster if you want multiple! model_1.save_pretrained_gguf( "llama_finetune", tokenizer_2, quantization_method="q4_k_m" ) if False: model_1.push_to_hub_gguf( "HF_USERNAME/llama_finetune", tokenizer_2, quantization_method="q4_k_m", token="YOUR_HF_TOKEN", # Get a token at https://huggingface.co/settings/tokens ) # Change hf to your username! if False: model_1.push_to_hub_gguf( "HF_USERNAME/llama_finetune", tokenizer_2, quantization_method=["q4_k_m", "q8_0", "q5_k_m"], token="YOUR_HF_TOKEN", # Get a token at https://huggingface.co/settings/tokens ) # Get a token at https://huggingface.co/settings/tokens return @app.cell(hide_code=True) def _(mo): mo.md(r""" And we're done! If you have any questions on Unsloth, we have a [Discord](https://discord.gg/unsloth) channel! If you find any bugs or want to keep updated with the latest LLM stuff, or need help, join projects etc, feel free to join our Discord! Some other resources: 1. Looking to use Unsloth locally? Read our [Installation Guide](https://unsloth.ai/docs/get-started/install) for details on installing Unsloth on Windows, Docker, AMD, Intel GPUs. 2. Learn how to do Reinforcement Learning with our [RL Guide and notebooks](https://unsloth.ai/docs/get-started/reinforcement-learning-rl-guide). 3. Read our guides and notebooks for [Text-to-speech (TTS)](https://unsloth.ai/docs/basics/text-to-speech-tts-fine-tuning) and [vision](https://unsloth.ai/docs/basics/vision-fine-tuning) model support. 4. Explore our [LLM Tutorials Directory](https://unsloth.ai/docs/models/tutorials-how-to-fine-tune-and-run-llms) to find dedicated guides for each model. 5. Need help with Inference? Read our [Inference & Deployment page](https://unsloth.ai/docs/basics/inference-and-deployment) for details on using vLLM, llama.cpp, Ollama etc.

Join Discord if you need help + ⭐️ Star us on Github ⭐️ This notebook and all Unsloth notebooks are licensed [LGPL-3.0](https://github.com/unslothai/notebooks?tab=LGPL-3.0-1-ov-file#readme)

""") return if __name__ == "__main__": app.run()