NeuralHermes-2.5-Mistral-7B

Description

This repo contains GGUF format model files for NeuralHermes-2.5-Mistral-7B.

Files Provided

Name	Quant	Bits	File Size	Remark
neuralhermes-2.5-mistral-7b.IQ3_S.gguf	IQ3_S	3	3.18 GB	3.44 bpw quantization
neuralhermes-2.5-mistral-7b.IQ3_M.gguf	IQ3_M	3	3.28 GB	3.66 bpw quantization mix
neuralhermes-2.5-mistral-7b.Q4_0.gguf	Q4_0	4	4.11 GB	3.56G, +0.2166 ppl
neuralhermes-2.5-mistral-7b.IQ4_NL.gguf	IQ4_NL	4	4.16 GB	4.25 bpw non-linear quantization
neuralhermes-2.5-mistral-7b.Q4_K_M.gguf	Q4_K_M	4	4.37 GB	3.80G, +0.0532 ppl
neuralhermes-2.5-mistral-7b.Q5_K_M.gguf	Q5_K_M	5	5.13 GB	4.45G, +0.0122 ppl
neuralhermes-2.5-mistral-7b.Q6_K.gguf	Q6_K	6	5.94 GB	5.15G, +0.0008 ppl
neuralhermes-2.5-mistral-7b.Q8_0.gguf	Q8_0	8	7.70 GB	6.70G, +0.0004 ppl

Parameters

path	type	architecture	rope_theta	sliding_win	max_pos_embed
teknium/OpenHermes-2.5-Mistral-7B	mistral	MistralForCausalLM	10000	4096	32768

Benchmarks

Original Model Card

NeuralHermes 2.5 - Mistral 7B

NeuralHermes is based on the teknium/OpenHermes-2.5-Mistral-7B model that has been further fine-tuned with Direct Preference Optimization (DPO) using the mlabonne/chatml_dpo_pairs dataset. It surpasses the original model on most benchmarks (see results).

It is directly inspired by the RLHF process described by Intel/neural-chat-7b-v3-1's authors to improve performance. I used the same dataset and reformatted it to apply the ChatML template.

The code to train this model is available on Google Colab and GitHub. It required an A100 GPU for about an hour.

Quantized models

Results

Update: NeuralHermes-2.5 became the best Hermes-based model on the Open LLM leaderboard and one of the very best 7b models. 🎉

Teknium (author of OpenHermes-2.5-Mistral-7B) benchmarked the model (see his tweet).

Results are improved on every benchmark: AGIEval (from 43.07% to 43.62%), GPT4All (from 73.12% to 73.25%), and TruthfulQA.

AGIEval

GPT4All

TruthfulQA

You can check the Weights & Biases project here.

Usage

You can run this model using LM Studio or any other frontend.

You can also run this model using the following code:

import transformers
from transformers import AutoTokenizer

# Format prompt
message = [
    {"role": "system", "content": "You are a helpful assistant chatbot."},
    {"role": "user", "content": "What is a Large Language Model?"}
]
tokenizer = AutoTokenizer.from_pretrained(new_model)
prompt = tokenizer.apply_chat_template(message, add_generation_prompt=True, tokenize=False)

# Create pipeline
pipeline = transformers.pipeline(
    "text-generation",
    model=new_model,
    tokenizer=tokenizer
)

# Generate text
sequences = pipeline(
    prompt,
    do_sample=True,
    temperature=0.7,
    top_p=0.9,
    num_return_sequences=1,
    max_length=200,
)
print(sequences[0]['generated_text'])

Training hyperparameters

LoRA:

r=16
lora_alpha=16
lora_dropout=0.05
bias="none"
task_type="CAUSAL_LM"
target_modules=['k_proj', 'gate_proj', 'v_proj', 'up_proj', 'q_proj', 'o_proj', 'down_proj']

Training arguments:

per_device_train_batch_size=4
gradient_accumulation_steps=4
gradient_checkpointing=True
learning_rate=5e-5
lr_scheduler_type="cosine"
max_steps=200
optim="paged_adamw_32bit"
warmup_steps=100

DPOTrainer:

beta=0.1
max_prompt_length=1024
max_length=1536

koesn
/

NeuralHermes-2.5-Mistral-7B-GGUF