Model version · Qwen3-4B-Base-Think-GRPO-ClipHigher0.28-KL0.001-LR1e-5-bf16-540steps
Qwen3-4B-Base-Think-GRPO-ClipHigher0.28-KL0.001-LR1e-5-bf16-540steps at 4176e078c77e
The Qwen3-4B-Base-Think-GRPO-ClipHigher0.28-KL0.001-LR1e-5-bf16-540steps repository at revision 4176e078c77e, as read on 2026-10-20.
This Revision
- Upstream revision
- 4176e078c77ea6d55f4a336a78f4f9df32e04eaa
- Observed by SAVRN
- 2026-10-20T03:47:39
- Artifact formats
- Not stated by the source
- Parameters, publisher reported
- Not stated by the source
- Architecture
- Not stated by the source
- License as stated
- Not stated by the source
- Released upstream
- 2026-10-20
All Recorded Versions
| Version | Observed | Formats | State |
|---|---|---|---|
| 4176e078c77e | 2026-10-20 | — | Current |