|
|
--- |
|
|
language: en |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- text-generation |
|
|
- question-answering |
|
|
- mcqa |
|
|
- merged |
|
|
- sft |
|
|
- lora |
|
|
base_model: AnnaelleMyriam/SFT_M3_model |
|
|
--- |
|
|
|
|
|
# MNLP M3 MCQA Merged Model |
|
|
|
|
|
This model is a merged version of: |
|
|
- **Base SFT Model**: `AnnaelleMyriam/SFT_M3_model` |
|
|
- **LoRA Adapter**: `aymanbakiri/MNLP_M3_mcqa_sft_model` |
|
|
|
|
|
## Model Description |
|
|
|
|
|
This is a specialized model for Multiple Choice Question Answering (MCQA) tasks, created by: |
|
|
1. Starting with the SFT model `AnnaelleMyriam/SFT_M3_model` |
|
|
2. Fine-tuning with LoRA adapters on MCQA data |
|
|
3. Merging the LoRA weights back into the base model |
|
|
|
|
|
## Usage |
|
|
|
|
|
```python |
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
|
|
model = AutoModelForCausalLM.from_pretrained("aymanbakiri/MNLP_M3_mcqa_merged_model") |
|
|
tokenizer = AutoTokenizer.from_pretrained("aymanbakiri/MNLP_M3_mcqa_merged_model") |
|
|
|
|
|
# Example usage for MCQA |
|
|
prompt = """Question: What is the capital of France? |
|
|
Options: (A) London (B) Berlin (C) Paris (D) Madrid |
|
|
Answer:""" |
|
|
|
|
|
inputs = tokenizer(prompt, return_tensors="pt") |
|
|
outputs = model.generate(**inputs, max_new_tokens=5) |
|
|
answer = tokenizer.decode(outputs[0], skip_special_tokens=True) |
|
|
print(answer) |
|
|
``` |
|
|
|
|
|
## Training Details |
|
|
|
|
|
- Base Model: SFT model fine-tuned for instruction following |
|
|
- LoRA Configuration: r=16, alpha=32, dropout=0.1 |
|
|
- Target Modules: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj, lm_head |
|
|
- Training Data: MNLP M2 MCQA Dataset |
|
|
|
|
|
## Performance |
|
|
|
|
|
This merged model should provide better performance than the original LoRA adapter while being easier to deploy and use. |
|
|
|
