ModelVerifiedExternal

Qwen3 4B Chemical GRPO TR

This repository contains the Qwen3-4B Chemical GRPO batch-size-32 run. The repository name uses the requested GRPO-TR suffix, but the actual training method for this checkpoint is GRPO.

Sourcehuggingface.co/SeongryongJung/Qwen3-4B-Chemical-GRPO-TR/tree/412f1c4d81b896853adc7bcecc637c1016187366

KindCheckpointBase modelQwen/Qwen3-4BVersionRevision 412f1c4Publisher@SeongryongJungCgrade

Model source

Kind: Checkpoint
File size: 3.58–4.64 GB
Base model: Qwen/Qwen3-4B
Version: vRevision 412f1c4
Updated: Oct 4, 2026
Publisher: @SeongryongJung
Downloads: 6
Parameters: 4B
Tasks: Text generation
Source: Hugging Face

Source description

Qwen3-4B-Chemical-GRPO-TR

This repository contains the Qwen3-4B Chemical GRPO batch-size-32 run. The repository name uses the requested GRPO-TR suffix, but the actual training method for this checkpoint is GRPO.

The repository root contains the best validation checkpoint (global_step_100), selected by validation mean@16. checkpoints/last/ contains the final checkpoint. For this run, best and final are the same checkpoint.

Performance

Dataset	Method	Base model	Train batch size	Best val mean@16	Best checkpoint	Final val mean@16	Final checkpoint
Chemical	GRPO	Qwen3-4B	32	68.81%	100	68.81%	100

Training and validation scores

Raw result files:

results/validation_mean16.csv
results/training_scores.csv
artifacts/config.yaml
artifacts/wandb-summary.json
artifacts/wandb-metadata.json
artifacts/output.log

Validation Mean@16

step	val_mean16
10	0.439880952381
20	0.505952380952
30	0.552678571429
40	0.590476190476
50	0.631845238095
60	0.648214285714
70	0.655952380952
80	0.665773809524
90	0.678869047619
100	0.688095238095

Training Hyperparameters

Hyperparameter	Value
Base model	`Qwen/Qwen3-4B`
Method	`GRPO`
Train batch size	`32`
Train max samples	`3200`
Total training steps	`100`
Save frequency	`10`
Test frequency	`10`
Rollout samples per prompt	`8`
Validation samples per prompt	`16`
Learning rate	`1e-6`
vLLM GPU memory utilization	`0.8`
Parsed train score rows	`100`

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

repo_id = "SeongryongJung/Qwen3-4B-Chemical-GRPO-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,
)

Source

Checkpoint: checkpoints/datasets/sciknoweval/chemistry/qwen3gen-chemistry-GRPO-Qwen-Qwen3-4B-mbs8-train32-rollout8-lr1e-6-vllm0.8
W&B run: run-20260702_050732-0ozhfamy
Full queue log: artifacts/queue.log

Sources

1 source

Hugging FacePrimaryVersion 412f1c4d81b896853adc7bcecc637c1016187366

huggingface.co/SeongryongJung/Qwen3-4B-Chemical-GRPO-TR/tree/412f1c4d81b896853adc7bcecc637c1016187366

Verified Oct 4

Download files

Model artifacts

2 artifacts

checkpoints/last/model-00001-of-00002.safetensors

safetensors · 4.64 GB · SHA-256 98c30e5afc7a…f585 · Hugging Face

Download shard

checkpoints/last/model-00002-of-00002.safetensors

safetensors · 3.58 GB · SHA-256 0b00d8a61e23…98c3 · Hugging Face

Download shard

Source excerpts

3 excerpts

ReadmeJul 11huggingface.co/SeongryongJung/Qwen3-4B-Chemical-GRPO-TR

--- license: apache-2.0 library_name: transformers pipeline_tag: text-generation tags: - qwen3 - reinforcement-learning - grpo - text-generation - text-generation-inference base_model: Qwen/Qwen3-4B --- # Qwen3-4B-Chemical-GRPO-TR This repository contains the Qwen3-4B `Chemical` `GRPO` batch-size-32 run. The repository name uses the requested `GRPO-TR` suffix, but the actual training method for this checkpoint is GRPO. The repository root contains the best validation checkpoint (`global_step_100`), selected by validation `mean@16`. `checkpoints/last/` contains the final checkpoint. For this run, best and final are the same checkpoint. ## Performance | Dataset | Method | Base model | Train batch size | Best val mean@16 | Best checkpoint | Final val mean@16 | Final checkpoint | |---|---|---|---:|---:|---:|---:|---:| | Chemical | GRPO | Qwen3-4B | 32 | 68.81% | 100 | 68.81% | 100 | ![Training and validation scores](results/training_score.png) Raw result files: - `results/validation_mean16.csv` - `results/training_scores.csv` - `artifacts/config.yaml` - `artifacts/wandb-summary.json` - `artifacts/wandb-metadata.json` - `artifacts/output.log` ## Validation Mean@16 | step | val_mean16 | |---:|---:| | 10 | 0.439880952381 | | 20 | 0.505952380952 | | 30 | 0.552678571429 | | 40 | 0.590476190476 | | 50 | 0.631845238095 | | 60 | 0.648214285714 | | 70 | 0.655952380952 | | 80 | 0.665773809524 | | 90 | 0.678869047619 | | 100 | 0.688095238095 | ## Training Hyperparameters | Hyperparameter | Value | |---|---| | Base model | `Qwen/Qwen3-4B` | | Method | `GRPO` | | Train batch size | `32` | | Train max samples | `3200` | | Total training steps | `100` | | Save frequency | `10` | | Test frequency | `10` | | Rollout samples per prompt | `8` | | Validation samples per prompt | `16` | | Learning rate | `1e-6` | | vLLM GPU memory utilization | `0.8` | | Parsed train score rows | `100` | ## Usage ```python from transformers import AutoModelForCausalLM, AutoTokenizer repo_id = "SeongryongJung/Qwen3-4B-Chemical-GRPO-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, ) ``` ## Source - Checkpoint: `checkpoints/datasets/sciknoweval/chemistry/qwen3gen-chemistry-GRPO-Qwen-Qwen3-4B-mbs8-train32-rollout8-lr1e-6-vllm0.8` - W&...

NoteOct 4huggingface.co/SeongryongJung/Qwen3-4B-Chemical-GRPO-TR

Source context: 6 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo SeongryongJung/Qwen3-4B-Chemical-GRPO-TR

NoteJul 18huggingface.co/SeongryongJung/Qwen3-4B-Chemical-GRPO-TR

Source context: 228 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo SeongryongJung/Qwen3-4B-Chemical-GRPO-TR