Qwens_SFT

This repository contains a model artifact fine-tuned from Qwen/Qwen3-2B-Instruct.

The model was trained to take a SMILES string and generate JSON with:

failure_summary_for_editing
failure_properties
failure_properties_display
editing_directions

Training data source: /home/onkarks2/transformers/fallen_angel_with_properties.csv

Example Prompt

Given the SMILES string below, generate the clinical failure editing information.

SMILES: Cc1cc(C)c(N)c(n1)C#Cc1cccc(F)c1

Return valid JSON with exactly these keys:
- "failure_summary_for_editing"
- "failure_properties"
- "failure_properties_display"
- "editing_directions"

Loading A LoRA Adapter

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base_model = "Qwen/Qwen3-2B-Instruct"
adapter_id = "onkarsus13/Qwens_SFT"

tokenizer = AutoTokenizer.from_pretrained(adapter_id)
model = AutoModelForCausalLM.from_pretrained(base_model, torch_dtype="auto", device_map="auto")
model = PeftModel.from_pretrained(model, adapter_id)

Downloads last month: -; Downloads are not tracked for this model. How to track