Qwen3-4B-Biology-SRPO-TR
This repository contains the Qwen3-4B Biology SRPO-TR run trained with batch size 32.
The repository root contains the best validation checkpoint (global_step_50), selected by validation mean@16.
checkpoints/last/ contains the final checkpoint.
Performance
| Dataset | Method | Base model | Train batch size | Best val mean@16 | Best checkpoint | Final val mean@16 | Final checkpoint |
|---|---|---|---|---|---|---|---|
| Biology | SRPO-TR | Qwen3-4B | 32 | 53.75% | 50 | 53.50% | 100 |
Raw result files:
results/validation_mean16.csvresults/training_scores.csvartifacts/config.yamlartifacts/wandb-summary.json
Training Hyperparameters
| Hyperparameter | Value |
|---|---|
| Base model | Qwen/Qwen3-4B |
| Method | SRPO-TR |
| Train batch size | 32 |
| Train max samples | 3200 |
| Total training steps | 100 |
| Save frequency | 10 |
| Test frequency | 10 |
| Teacher regularization | trust-region |
| Teacher update rate | 0.1 |
| Rollout samples per prompt | 8 |
| Validation samples per prompt | 16 |
| Learning rate | 5e-6 |
| vLLM GPU memory utilization | 0.8 |
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
repo_id = "SeongryongJung/Qwen3-4B-Biology-SRPO-TR"
tokenizer = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
repo_id,
torch_dtype="auto",
device_map="auto",
trust_remote_code=True,
)