You can deploy this model and start earning money today!

RichardErkhov/
mrm8488_-_mistral-7b-ft-h4-no_robots_instructions-gguf

Name: mrm8488_-_mistral-7b-ft-h4-no_robots_instructions-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mistral-7b-ft-h4-no_robots_instructions - GGUF

Model creator: https://huggingface.co/mrm8488/
Original model: https://huggingface.co/mrm8488/mistral-7b-ft-h4-no_robots_instructions/

Name	Quant method	Size
mistral-7b-ft-h4-no_robots_instructions.Q2_K.gguf	Q2_K	2.53GB
mistral-7b-ft-h4-no_robots_instructions.IQ3_XS.gguf	IQ3_XS	2.81GB
mistral-7b-ft-h4-no_robots_instructions.IQ3_S.gguf	IQ3_S	2.96GB
mistral-7b-ft-h4-no_robots_instructions.Q3_K_S.gguf	Q3_K_S	2.95GB
mistral-7b-ft-h4-no_robots_instructions.IQ3_M.gguf	IQ3_M	3.06GB
mistral-7b-ft-h4-no_robots_instructions.Q3_K.gguf	Q3_K	3.28GB
mistral-7b-ft-h4-no_robots_instructions.Q3_K_M.gguf	Q3_K_M	3.28GB
mistral-7b-ft-h4-no_robots_instructions.Q3_K_L.gguf	Q3_K_L	3.56GB
mistral-7b-ft-h4-no_robots_instructions.IQ4_XS.gguf	IQ4_XS	3.38GB
mistral-7b-ft-h4-no_robots_instructions.Q4_0.gguf	Q4_0	3.83GB
mistral-7b-ft-h4-no_robots_instructions.IQ4_NL.gguf	IQ4_NL	3.87GB
mistral-7b-ft-h4-no_robots_instructions.Q4_K_S.gguf	Q4_K_S	3.86GB
mistral-7b-ft-h4-no_robots_instructions.Q4_K.gguf	Q4_K	4.07GB
mistral-7b-ft-h4-no_robots_instructions.Q4_K_M.gguf	Q4_K_M	4.07GB
mistral-7b-ft-h4-no_robots_instructions.Q4_1.gguf	Q4_1	3.42GB
mistral-7b-ft-h4-no_robots_instructions.Q5_0.gguf	Q5_0	4.65GB
mistral-7b-ft-h4-no_robots_instructions.Q5_K_S.gguf	Q5_K_S	4.65GB
mistral-7b-ft-h4-no_robots_instructions.Q5_K.gguf	Q5_K	4.78GB
mistral-7b-ft-h4-no_robots_instructions.Q5_K_M.gguf	Q5_K_M	4.78GB
mistral-7b-ft-h4-no_robots_instructions.Q5_1.gguf	Q5_1	5.07GB
mistral-7b-ft-h4-no_robots_instructions.Q6_K.gguf	Q6_K	5.53GB
mistral-7b-ft-h4-no_robots_instructions.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0 datasets:

HuggingFaceH4/no_robots base_model: mistralai/Mistral-7B-v0.1 language:
en pipeline_tag: text-generation thumbnail: https://huggingface.co/mrm8488/mistral-7b-ft-h4-no_robots_instructions/resolve/main/mistralh4-removebg-preview.png?download=true

limstral logo

Mistral 7B fine-tuned on H4/No Robots instructions

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the HuggingFaceH4/no_robots dataset for instruction following downstream task.

Training procedure

The model was loaded on 8 bits and fine-tuned on the LIMA dataset using the LoRA PEFT technique with the huggingface/peft library and trl/sft for one epoch on 1 x A100 (40GB) GPU.

SFT Trainer params:

trainer = SFTTrainer(
    model=model,
    train_dataset=train_ds,
    eval_dataset=test_ds,
    peft_config=peft_config,
    dataset_text_field="text",
    max_seq_length=2048,
    tokenizer=tokenizer,
    args=training_arguments,
    packing=False
)

LoRA config:

config = LoraConfig(
        lora_alpha=16,
        lora_dropout=0.1,
        r=64,
        bias="none",
        task_type="CAUSAL_LM",
        target_modules = ['q_proj', 'k_proj', 'down_proj', 'v_proj', 'o_proj', 'gate_proj', 'up_proj']
    )

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 2
eval_batch_size: 8
seed: 66
gradient_accumulation_steps: 64
total_train_batch_size: 128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.03
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Step	Training Loss	Validation Loss
10	1.796200	1.774305
20	1.769700	1.679720
30	1.626800	1.667754
40	1.663400	1.665188
50	1.565700	1.659000
60	1.660300	1.658270

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

repo_id = "mrm8488/mistral-7b-ft-h4-no_robots_instructions"

model = AutoModelForCausalLM.from_pretrained(repo_id, torch_dtype=torch.bfloat16)
tokenizer = AutoTokenizer.from_pretrained(repo_id)

gen = pipeline("text-generation", model=model, tokenizer=tokenizer, device=0)

instruction = "[INST] Write an email to say goodbye to me boss [\INST]"
res = gen(instruction, max_new_tokens=512, temperature=0.3, top_p=0.75, top_k=40, repetition_penalty=1.2, eos_token_id=2)
print(res[0]['generated_text'])

Framework versions

Transformers 4.35.0.dev0
Pytorch 2.1.0+cu118
Datasets 2.14.6
Tokenizers 0.14.1

Citation

@misc {manuel_romero_2023,
	author       = { {Manuel Romero} },
	title        = { mistral-7b-ft-h4-no_robots_instructions (Revision 785446d) },
	year         = 2023,
	url          = { https://huggingface.co/mrm8488/mistral-7b-ft-h4-no_robots_instructions },
	doi          = { 10.57967/hf/1426 },
	publisher    = { Hugging Face }
}

RichardErkhov/mrm8488_-_mistral-7b-ft-h4-no_robots_instructions-gguf