qwen3-1.7b-a8_klreward-krew-k1-coef1e-3-mb4-eta100-bs256x5-n2

RL fine-tuned checkpoint from the KL-in-LLM-RL / FRPO experiments (trained with verl).

  • Base model: Qwen/Qwen3-1.7B
  • Checkpoint(s) in this repo: global_step_200 (repo root)
  • Weights: fp32 safetensors, exactly as saved by the trainer (no post-processing).
  • Run configuration is encoded in the repo name.

Auto-uploaded on 2026-08-15.

Downloads last month
14
Safetensors
Model size
2B params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for FRPO/qwen3-1.7b-a8_klreward-krew-k1-coef1e-3-mb4-eta100-bs256x5-n2

Finetuned
Qwen/Qwen3-1.7B
Finetuned
(1,416)
this model