Downloads · 30 days
20
9% of all-time downloads
flawed-fictions-gemma-3-4b
agurung/flawed-fictions-gemma-3-4b
flawed-fictions-gemma-3-4b 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
All-time downloads
218
Public
Parameters
4.3B
199 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors9.9 GB · 100%
At a glance
- Task
- Reinforcement Learning
- License
- apache-2.0
- Model type
- gemma3
- Access
- Public
- Created
- Feb 15, 2026
- Updated
- Oct 1, 2026
- SHA
- 1864f221
Base models
Datasets
Model card
From the Hugging Face model README
Flawed Fictions GRPO (gemma)
Training Details
| Base model | google/gemma-3-4b-it |
| Task | Continuity error detection (\boxed{Yes} / \boxed{No}) |
| W&B group | grpo_flawed_fictions_gemma |
| W&B runs | 30a1oo7m, kfqu25ha |
| Training script | scripts/grpo_4gpu_gemma_train.sh |
Checkpoint Revisions
- Branch head (latest):
main - Per-checkpoint tags:
main-step-<N>
Usage
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained(
"agurung/flawed-fictions-gemma-3-4b",
revision="main",
device_map="auto",
torch_dtype="auto",
)
- Task
- Reinforcement Learning
- Type
- gemma3
- License
- apache-2.0
- Created
- Feb 15, 2026
- Updated
- Oct 1, 2026