Description model

Chocolatine-3B version specialized in French administrative language, supervised fine-tuning of microsoft/Phi-3.5-mini-instruct
based on the official lexicon
published by the French Ministère de la Fonction Publique et de la Réforme de l'Etat.

Data & Training

The dataset gathers 2362 administrative terms constituting the basis of the simulation of prompt-answer pairs.
The GPT-4o model deployed on Azure OpenAI was used to carry out the building of the dataset in several phases:

Extraction of the lexicon pages converted into jpg images
Reformulation of the definitions to make them more readable and natural to be used by an LLM in order to ensure high quality data.
Generation of questions from the terms and definitions
Generation of answers in three successive rounds taking into account the previous generations to ensure variety.

Fine tuning (SFT) done efficiently with Unsloth,
with which I saved processing time on a single T4 GPU (Compute instance from Azure ML).

Usage

The recommended usage is by loading the low-rank adapter using unsloth:

from unsloth import FastLanguageModel

model_name = "jpacifico/chocolatine-admin-3B-sft-v0.2"
model, tokenizer = FastLanguageModel.from_pretrained(
  model_name = model_name,
  max_seq_length = 2048,
  dtype = None,
  load_in_4bit = True,
)

FastLanguageModel.for_inference(model)

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
License: MIT
Finetuned from model : microsoft/Phi-3.5-mini-instruct