Downloads · 30 days
13
4% of all-time downloads
agurung/flawed-fictions-qwen25-7b-lengthpenalty
flawed-fictions-qwen25-7b-lengthpenalty is a reinforcement learning model from agurung. 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.
- Branch head (latest): main - Per-checkpoint tags: main-step-<N
Downloads · 30 days
13
4% of all-time downloads
All-time downloads
309
Public
Parameters
7.6B
289 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors15.2 GB · 100%
From the Hugging Face model README
| Base model | Qwen/Qwen2.5-7B-Instruct |
| Task | Continuity error detection (\boxed{Yes} / \boxed{No}) |
| W&B group | flawed_fictions_rl |
| W&B runs | vttgr0i0 |
| Training script | scripts/grpo_4h100_qwen_lengthpenalty_train.sh |
mainmain-step-<N>from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained(
"agurung/flawed-fictions-qwen25-7b-lengthpenalty",
revision="main",
device_map="auto",
torch_dtype="auto",
)