Downloads · 30 days
10
4% of all-time downloads
Qwen3-4B-Material-RLSD-TR
SeongryongJung/Qwen3-4B-Material-RLSD-TR
Qwen3-4B-Material-RLSD-TR is a text generation model from SeongryongJung. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This repository contains the Qwen3-4B material RLSDTR batch-size-32 run.
All-time downloads
257
Public
Parameters
4.4B
8.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors17.6 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- apache-2.0
- Model type
- qwen3
- Access
- Public
- Created
- Jul 3, 2026
- Updated
- Jul 3, 2026
- SHA
- 6b5b1c1e
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3-4B-Material-RLSD-TR
This repository contains the Qwen3-4B material RLSD_TR batch-size-32 run.
The repository root contains the best validation checkpoint, 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 |
|---|---|---|---|---|---|---|---|
| Material / SciKnowEval material | RLSD_TR | Qwen3-4B | 32 | 76.86% | 100 | 76.86% | 100 |
Validation Mean@16
| step | val_mean16 | percent |
|---|---|---|
| 10 | 0.652925531915 | 65.29% |
| 20 | 0.701462765957 | 70.15% |
| 30 | 0.725398936170 | 72.54% |
| 40 | 0.742021276596 | 74.20% |
| 50 | 0.747340425532 | 74.73% |
| 60 | 0.760638297872 | 76.06% |
| 70 | 0.766622340426 | 76.66% |
| 80 | 0.757978723404 | 75.80% |
| 90 | 0.763962765957 | 76.40% |
| 100 | 0.768617021277 | 76.86% |
Detailed Training Hyperparameters
| Section | Parameter | Value | Source |
|---|---|---|---|
| Run identity | Base model | Qwen/Qwen3-4B | queue/script override |
| Run identity | Dataset | Material / SciKnowEval material | run_qwen3_generalization.sh |
| Run identity | Method | RLSD_TR | run_qwen3_generalization.sh |
| Run identity | Config | rlsd | run_qwen3_generalization.sh |
| Run identity | Experiment | qwen3gen-material-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8 | run_qwen3_generalization.sh |
| Run identity | W&B run | run-20260702_135403-jd6eugui | wandb |
| Data | Train file | datasets/sciknoweval/material/train.parquet | script override |
| Data | Validation file | datasets/sciknoweval/material/test.parquet | script override |
| Data | Train batch size | 32 | queue/script override |
| Data | Train max samples | 3200 | queue/script override |
| Schedule | Total training steps | 100 | queue/script override |
| Schedule | Validation before train | False | queue/script override |
| Schedule | Save frequency | 10 | queue/script override |
| Schedule | Validation frequency | 10 | queue/script override |
| Sequence | Max prompt length | 2048 | queue/script override |
| Sequence | Max response length | 8192 | queue/script override |
| Sequence | Max model length | 10240 | queue/script override |
| Rollout | Train rollout n | 8 | queue/script override |
| Rollout | Validation rollout n | 16 | queue/script override |
| Rollout | vLLM GPU memory utilization | 0.8 | queue/script override |
| Optimization | Learning rate | 1e-6 | RLSD_TR method override |
| Optimization | Weight decay | 0.01 | script override |
| PPO/GRPO | PPO mini batch size | 8 | queue/script override |
| PPO/GRPO | Normalize GRPO advantages by std | False | baseline_grpo.yaml / script override |
| Rollout correction | Importance sampling mode | token | script override |
| Rollout correction | IS threshold | 2.0 | script override |
| Checkpoint/Logging | Checkpoint root | checkpoints/datasets/sciknoweval/material/qwen3gen-material-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8 | script override |
| Checkpoint/Logging | Latest checkpointed iteration | 100 | latest_checkpointed_iteration.txt |
| Checkpoint/Logging | External actor archive | checkpoints/datasets/sciknoweval/material/qwen3gen-material-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/_actor_archive | preserve_actor_checkpoints.py |
| Checkpoint/Logging | Logger | console, wandb | ppo_trainer.yaml |
| RLSD_TR | Policy loss mode | rlsd | method override |
| RLSD_TR | Teacher regularization | trust-region | method override |
| RLSD_TR | Trust-region mix / teacher update rate | 0.1 | queue/script override |
| RLSD_TR | Token reweight lambda | 0.5 | queue/script override |
| RLSD_TR | Token reweight eps_w | 0.2 | queue/script override |
| RLSD_TR | Token reweight decay steps | 0 | queue/script override |
| RLSD_TR | Fused kernels | False | method override |
Raw result and artifact files:
results/validation_mean16.csvresults/training_scores.csvresults/hyperparameters.csvresults/training_score.pngresults/training_score.svgartifacts/config.yamlartifacts/wandb-summary.jsonartifacts/wandb-metadata.jsonartifacts/output.logartifacts/queue.log
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
repo_id = "SeongryongJung/Qwen3-4B-Material-RLSD-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/material/qwen3gen-material-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8 - Root actor checkpoint:
checkpoints/datasets/sciknoweval/material/qwen3gen-material-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/global_step_100/actor - Last actor checkpoint:
checkpoints/datasets/sciknoweval/material/qwen3gen-material-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/global_step_100/actor - W&B run:
run-20260702_135403-jd6eugui - Queue log:
artifacts/queue.log
- Task
- Text Generation
- Library
- transformers
- Type
- qwen3
- License
- apache-2.0
- Created
- Jul 3, 2026
- Updated
- Jul 3, 2026