Downloads · 30 days
18
24% of all-time downloads
stanleytheli/qwen3.6-35B-A3B-av-RL1200
qwen3.6-35B-A3B-av-RL1200 is a reinforcement learning model from stanleytheli. 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.
RL-tuned Activation Verbalizer (AV) for Qwen/Qwen3.6-35B-A3B — the SFT model qwen3.6-35B-A3B-av-sft after 1200 steps of GRPO against a frozen reconstruction reward. Companion reward/ reconstruction model (frozen throu…
Downloads · 30 days
18
24% of all-time downloads
All-time downloads
76
Public
Parameters
34.7B
69.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors69.3 GB · 100%
From the Hugging Face model README
RL-tuned Activation Verbalizer (AV) for Qwen/Qwen3.6-35B-A3B — the SFT model
qwen3.6-35B-A3B-av-sft
after 1200 steps of GRPO against a frozen reconstruction reward. Companion reward/
reconstruction model (frozen throughout RL):
qwen3.6-35B-A3B-ar-sft.
| checkpoint | FVE (norm) | cos mean / median | FVE (raw) |
|---|---|---|---|
| SFT (av-sft) | 0.418 | 0.807 / 0.835 | −0.26 |
| GRPO 200 | 0.583 | 0.862 / 0.887 | +0.17 |
| GRPO 1200 (this) | 0.644 | 0.882 / 0.907 | +0.32 |
Same injection contract as the SFT model: nla_meta.yaml + injection_scale.txt
included; scan for the marker token id at inference time (never precompute
positions). See the NLA repo for injection utilities.