RichardErkhov/HuggingFaceH4_-_mistral-7b-sft-beta-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mistral-7b-sft-beta - GGUF

• Model creator: https://huggingface.co/HuggingFaceH4/
• Original model: https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta/

Name	Quant method	Size
mistral-7b-sft-beta.Q2_K.gguf	Q2_K	2.53GB
mistral-7b-sft-beta.IQ3_XS.gguf	IQ3_XS	2.81GB
mistral-7b-sft-beta.IQ3_S.gguf	IQ3_S	2.96GB
mistral-7b-sft-beta.Q3_K_S.gguf	Q3_K_S	2.95GB
mistral-7b-sft-beta.IQ3_M.gguf	IQ3_M	3.06GB
mistral-7b-sft-beta.Q3_K.gguf	Q3_K	3.28GB
mistral-7b-sft-beta.Q3_K_M.gguf	Q3_K_M	3.28GB
mistral-7b-sft-beta.Q3_K_L.gguf	Q3_K_L	3.56GB
mistral-7b-sft-beta.IQ4_XS.gguf	IQ4_XS	3.67GB
mistral-7b-sft-beta.Q4_0.gguf	Q4_0	3.83GB
mistral-7b-sft-beta.IQ4_NL.gguf	IQ4_NL	3.87GB
mistral-7b-sft-beta.Q4_K_S.gguf	Q4_K_S	3.86GB
mistral-7b-sft-beta.Q4_K.gguf	Q4_K	4.07GB
mistral-7b-sft-beta.Q4_K_M.gguf	Q4_K_M	4.07GB
mistral-7b-sft-beta.Q4_1.gguf	Q4_1	4.24GB
mistral-7b-sft-beta.Q5_0.gguf	Q5_0	4.65GB
mistral-7b-sft-beta.Q5_K_S.gguf	Q5_K_S	4.65GB
mistral-7b-sft-beta.Q5_K.gguf	Q5_K	4.78GB
mistral-7b-sft-beta.Q5_K_M.gguf	Q5_K_M	4.78GB
mistral-7b-sft-beta.Q5_1.gguf	Q5_1	5.07GB
mistral-7b-sft-beta.Q6_K.gguf	Q6_K	5.53GB

Original model description:

license: mit base_model: mistralai/Mistral-7B-v0.1 tags: - generated_from_trainer model-index: - name: mistral-7b-sft-beta results: [] datasets: - HuggingFaceH4/ultrachat_200k language: - en

Model Card for Mistral 7B SFT β

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the HuggingFaceH4/ultrachat_200k dataset. It is the SFT model that was used to train Zephyr-7B-β with Direct Preference Optimization.

It achieves the following results on the evaluation set:

• Loss: 0.9399

Model description

• Model type: A 7B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
• Language(s) (NLP): Primarily English
• License: MIT
• Finetuned from model: mistralai/Mistral-7B-v0.1

Model Sources

• Repository: https://github.com/huggingface/alignment-handbook

Intended uses & limitations

The model was fine-tuned with 🤗 TRL's SFTTrainer on a filtered and preprocessed of the UltraChat dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT.

Here's how you can run the model using the pipeline() function from 🤗 Transformers:

# Install transformers from source - only needed for versions <= v4.34
# pip install git+https://github.com/huggingface/transformers.git
# pip install accelerate

import torch
from transformers import pipeline

pipe = pipeline("text-generation", model="HuggingFaceH4/mistral-7b-sft-beta", torch_dtype=torch.bfloat16, device_map="auto")

# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
messages = [
    {
        "role": "system",
        "content": "You are a friendly chatbot who always responds in the style of a pirate",
    },
    {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
# <|system|>
# You are a friendly chatbot who always responds in the style of a pirate.</s>
# <|user|>
# How many helicopters can a human eat in one sitting?</s>
# <|assistant|>
# Ah, me hearty matey! But yer question be a puzzler! A human cannot eat a helicopter in one sitting, as helicopters are not edible. They be made of metal, plastic, and other materials, not food!

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• distributed_type: multi-GPU
• num_devices: 16
• gradient_accumulation_steps: 4
• total_train_batch_size: 512
• total_eval_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.9367	0.67	272	0.9397

Framework versions

• Transformers 4.35.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.14.0