Vous avez un workflow ComfyUI ? Obtenez une documentation lisible directement dans votre navigateur.Rendre un workflow lisible.Essayer maintenant ↗
license: apache-2.0 libraryname: transformers pipelinetag: text-generation tags: qwen3 reinforcement-learning grpo text-generation text-generation-inference basemodel: Qwen/Qwen3-4B
Source du modèle
Description de la source
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 (), selected by validation . contains the final checkpoint. For this run, best and final are the same checkpoint.
Sources
1 sourceVérifié 18 juil.
Artefacts du modèle
2 artefactscheckpoints/last/model-00001-of-00002.safetensors
safetensors · 4,64 GB · SHA-256 98c30e5afc7a…f585 · Hugging Face
Téléchargermodel-00001-of-00002.safetensors
safetensors · 4,64 GB · SHA-256 98c30e5afc7a…f585 · Hugging Face
TéléchargerExtraits de sources
3 extraits
global_step_100mean@16checkpoints/last/| 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.csvresults/training_scores.csvartifacts/config.yamlartifacts/wandb-summary.jsonartifacts/wandb-metadata.jsonartifacts/output.log| 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 |
| 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 |
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,
)
checkpoints/datasets/sciknoweval/chemistry/qwen3gen-chemistry-GRPO-Qwen-Qwen3-4B-mbs8-train32-rollout8-lr1e-6-vllm0.8run-20260702_050732-0ozhfamyartifacts/queue.log--- 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 |  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&...
Source context: 228 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo SeongryongJung/Qwen3-4B-Chemical-GRPO-TR
Source context: 221 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo SeongryongJung/Qwen3-4B-Chemical-GRPO-TR
0.8 |
| Parsed train score rows | 100 |