Edit model card

Fine-tuning Phi-2 Model on Medical Dataset

😸GitHub •📝Article • Models & Datasets on: 🤗Hugging Face

diagram

This repository provides a collection of Jupyter notebooks that demonstrate how to fine-tune large language models using various tools and techniques.

fine-tuning or instruction tuning is the process where the pre-trained model is further trained on the smaller dataset to adapt its knowledge for a specific task or domain. This process tweaks the model’s parameters to perform specific tasks. In fine-tuning, there are two methods:

Supervised fine tuning (SFT):

In SFT, the model is trained on a labeled dataset. The labeled dataset typically contains examples of instruction (input) and response (output) pairs relevant to the task. In this process, the model learns how to respond to specific instructions.

Reinforcement Learning from Human Feedback (RLHF):

In RLHF, the model interacts with users, generates responses, and receives feedback in the form of reinforcement signals. Basically, the model learns and improves its performance based on the feedback it receives.

Details of each notebook:

LLMs Description Dataset Notebooks Tuned Model
Phi-2 This model has been fine-tuned on a Medical dataset to answer questions related to diseases and symptoms. (used SFT method) Dataset Open In Colab phi-2-medquad
llama-2 This model has been fine-tuned on a dataset with human-generated prompts to answer questions related to general knowledge. (used SFT method) Dataset Open In Colab llama-2-finance
Downloads last month
32
Safetensors
Model size
2.78B params
Tensor type
FP16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for prsdm/phi-2-medquad

Base model

microsoft/phi-2
Finetuned
(284)
this model

Dataset used to train prsdm/phi-2-medquad