ealharbi/crystallography-cryoem-qwen2.5-72b-lora

Qwen2.5 72B — crystallography LoRA adapter

LoRA adapter for Qwen/Qwen2.5-72B-Instruct, fine-tuned on confirmed question-and-answer pairs from the CCP4BB, CCP-EM, phenixbb and COOT mailing lists.

Training

Method	LoRA, r=64, α=128, dropout=0.05
Learning rate	2.675e-05
Max length	2048 tokens
Epochs	2, early stopping
Seed	1

Evaluation

BERTScore F1 against held-out confirmed answers (433 questions, roberta-large, score(prediction, reference)): 0.8212 → 0.8542.

BERTScore measures overlap with the reference wording, not factual correctness; the paper reports three LLM judges alongside it.

Use

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-72B-Instruct")
model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen2.5-72B-Instruct", device_map="auto", dtype="bfloat16")
model = PeftModel.from_pretrained(model, "ealharbi/crystallography-cryoem-qwen2.5-72b-lora")

The system prompt used in training was:

You are a helpful crystallography assistant. Be concise and precise.

The training pairs are single-turn, so ask one self-contained question at a time; multi-turn prompts fall outside the fine-tuning format.

Caveat

Answers are generated and may be wrong. Verify anything consequential against the program documentation and the primary literature.