ThakrePranjal/pharma-tinyllama-domain-lora

Pharma TinyLlama — Domain LoRA Adapter

This is a LoRA adapter for TinyLlama/TinyLlama-1.1B-Chat-v1.0, trained via domain-adaptive continued pretraining on a pharmaceutical text corpus.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model_name = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"
adapter_repo = "ThakrePranjal/pharma-tinyllama-domain-lora"

base_model = AutoModelForCausalLM.from_pretrained(base_model_name)
model = PeftModel.from_pretrained(base_model, adapter_repo)
tokenizer = AutoTokenizer.from_pretrained(adapter_repo)

Training data

Trained on ThakrePranjal/pharma-domain-corpus, a cleaned pharmaceutical text corpus.

Intended use

Continued pretraining / domain adaptation for pharma-domain text generation. This is not an instruction-tuned model — it continues text rather than following instructions. See the companion instruction-tuning notebook for an instruction-following variant.

Limitations

Trained on a small/sample pharma corpus; not validated for clinical or production use. Outputs should be verified against authoritative sources.