README.md

---
base_model: google/gemma-4-E2B-it
library_name: peft
tags:
- lora
- jmh
- rejection-fine-tuning
- mutation-testing
- gemma4
- text-generation
license: gemma
pipeline_tag: text-generation
---

# Gemma 4 E2B RFT LoRA — fastutil (mutation)

LoRA adapter from **rejection fine-tuning (RFT)** on `google/gemma-4-E2B-it` for JMH benchmark generation on **fastutil** classes with performance-mutation rewards.

## Training summary

| | |
|---|---|
| Base model | [google/gemma-4-E2B-it](https://huggingface.co/google/gemma-4-E2B-it) |
| Method | LoRA (r=16, alpha=32) + bf16, 1 epoch SFT on accepted RFT traces |
| Corpus | 14 mutation-scored fastutil classes |
| Train samples | 15 (+ 1 val) |
| Accepted / generated | 17 / 224 (7.6%) |
| Max seq len | 16384 (chunked CE loss) |

## Load and use

```python
import torch
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = "google/gemma-4-E2B-it"
adapter = "bookxd/gemma-4-e2b-rft-fastutil-mutation"

tokenizer = AutoTokenizer.from_pretrained(adapter)
model = AutoModelForCausalLM.from_pretrained(
    base,
    torch_dtype=torch.bfloat16,
    attn_implementation="sdpa",
    device_map="auto",
)
model = PeftModel.from_pretrained(model, adapter)
model.eval()

messages = [
    {"role": "system", "content": "You write JMH benchmarks..."},
    {"role": "user", "content": "Target class: it.unimi.dsi.fastutil.ints.IntArrayList\n..."},
]
inputs = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt",
    chat_template_kwargs={"enable_thinking": True},
).to(model.device)

with torch.no_grad():
    out = model.generate(**inputs, max_new_tokens=8192, do_sample=True, temperature=1.0)
print(tokenizer.decode(out[0], skip_special_tokens=False))
```

## Files

- `adapter_model.safetensors` — LoRA weights (~92M params trainable on base)
- `adapter_config.json` — PEFT config (base model + target modules)
- `tokenizer.json`, `tokenizer_config.json`, `chat_template.jinja` — tokenizer + Gemma 4 thinking template

## Framework versions

- PEFT 0.19.1, TRL 1.5.1, Transformers 5.10.1, PyTorch 2.12.1