Downloads · 30 days
6
7% of all-time downloads
GRPO-TCR-Qwen3-4B-True-epoch2
y-ohtani/GRPO-TCR-Qwen3-4B-True-epoch2
GRPO-TCR-Qwen3-4B-True-epoch2 is a reinforcement learning model from y-ohtani. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
GRPO-TCR-Qwen3-4B-True epoch 2 (step 4000). Trained with nrespperprompt=8, trainbatch=4, SFT epoch3 base, maxresponselength=10480.
All-time downloads
84
Public
Parameters
4.4B
8.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors8.8 GB · 100%
At a glance
- Task
- Reinforcement Learning
- License
- apache-2.0
- Model type
- qwen3
- Access
- Public
- Created
- Mar 7, 2026
- Updated
- Mar 7, 2026
- SHA
- 45257cf5
Base models
Datasets
Model card
From the Hugging Face model README
GRPO-TCR-Qwen3-4B-True-epoch2
GRPO-TCR-Qwen3-4B-True epoch 2 (step 4000). Trained with n_resp_per_prompt=8, train_batch=4, SFT epoch3 base, max_response_length=10480.
Model Details
- Base Model: Qwen/Qwen3-4B-Instruct-2507
- Method: GRPO (Agentic RL)
- Dataset: y-ohtani/open_agentrl_30k_mix
Usage
This model is compatible with the transformers library.
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "y-ohtani/GRPO-TCR-Qwen3-4B-True-epoch2"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)
- Task
- Reinforcement Learning
- Type
- qwen3
- License
- apache-2.0
- Created
- Mar 7, 2026
- Updated
- Mar 7, 2026