DPO fine-tuned of microsoft/Phi-3.5-mini-instruct (3.82B params)
using the jpacifico/french-orca-dpo-pairs-revised rlhf dataset.
Training in French also improves the model in English, surpassing the performances of its base model.

MT-Bench-French

Chocolatine-3B-Instruct-DPO-v1.2 is outperforming Phi-3-medium-4k-instruct (14B) and its base model Phi-3.5-mini-instruct on MT-Bench-French, used with multilingual-mt-bench and GPT-4-Turbo as LLM-judge.

########## First turn ##########
                                             score
model                                 turn        
gpt-4o-mini                           1     9.2875
Chocolatine-14B-Instruct-4k-DPO       1     8.6375
Chocolatine-14B-Instruct-DPO-v1.2     1     8.6125
Phi-3.5-mini-instruct                 1     8.5250
Chocolatine-3B-Instruct-DPO-v1.2      1     8.3750
Phi-3-medium-4k-instruct              1     8.2250
gpt-3.5-turbo                         1     8.1375
Chocolatine-3B-Instruct-DPO-Revised   1     7.9875
Daredevil-8B                          1     7.8875
Meta-Llama-3.1-8B-Instruct            1     7.0500
vigostral-7b-chat                     1     6.7875
Mistral-7B-Instruct-v0.3              1     6.7500
gemma-2-2b-it                         1     6.4500
French-Alpaca-7B-Instruct_beta        1     5.6875
vigogne-2-7b-chat                     1     5.6625

########## Second turn ##########
                                               score
model                                 turn          
gpt-4o-mini                           2     8.912500
Chocolatine-14B-Instruct-DPO-v1.2     2     8.337500
Chocolatine-3B-Instruct-DPO-Revised   2     7.937500
Chocolatine-3B-Instruct-DPO-v1.2      2     7.862500
Phi-3-medium-4k-instruct              2     7.750000
Chocolatine-14B-Instruct-4k-DPO       2     7.737500
gpt-3.5-turbo                         2     7.679167
Phi-3.5-mini-instruct                 2     7.575000
Daredevil-8B                          2     7.087500
Meta-Llama-3.1-8B-Instruct            2     6.787500
Mistral-7B-Instruct-v0.3              2     6.500000
vigostral-7b-chat                     2     6.162500
gemma-2-2b-it                         2     6.100000
French-Alpaca-7B-Instruct_beta        2     5.487395
vigogne-2-7b-chat                     2     2.775000

########## Average ##########
                                          score
model                                          
gpt-4o-mini                            9.100000
Chocolatine-14B-Instruct-DPO-v1.2      8.475000
Chocolatine-14B-Instruct-4k-DPO        8.187500
Chocolatine-3B-Instruct-DPO-v1.2       8.118750
Phi-3.5-mini-instruct                  8.050000
Phi-3-medium-4k-instruct               7.987500
Chocolatine-3B-Instruct-DPO-Revised    7.962500
gpt-3.5-turbo                          7.908333
Daredevil-8B                           7.487500
Meta-Llama-3.1-8B-Instruct             6.918750
Mistral-7B-Instruct-v0.3               6.625000
vigostral-7b-chat                      6.475000
gemma-2-2b-it                          6.275000
French-Alpaca-7B-Instruct_beta         5.587866
vigogne-2-7b-chat                      4.218750

Usage

You can run this model using my Colab notebook

You can also run Chocolatine using the following code:

python

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'])

4-bit quantized version is available here : jpacifico/Chocolatine-3B-Instruct-DPO-v1.2-Q4_K_M-GGUF

Limitations

The Chocolatine model is a quick demonstration that a base model can be easily fine-tuned to achieve compelling performance.
It does not have any moderation mechanism.

Developed by: Jonathan Pacifico, 2024
Model type: LLM
Language(s) (NLP): French, English
License: MIT

RichardErkhov/jpacifico_-_Chocolatine-3B-Instruct-DPO-v1.2-8bits

Original model description:

Chocolatine-3B-Instruct-DPO-v1.2

MT-Bench-French

Usage

Limitations

RichardErkhov/
jpacifico_-_Chocolatine-3B-Instruct-DPO-v1.2-8bits