CoolFace
Modelpublic

Yooniel/qwen2.5-7b-instruct-nla-L20

sourceHugging Faceapache-2.0updated 11d agoView on Hugging Face
0likes
14 commits on main
8b35afe11d ago

Upload ckpts25_cold_steg_kl0_rawfallback checkpoints

Yooniel
3fd969011d ago

Upload ckpts25_cold_origprompt_kl0_rawfallback checkpoints

Yooniel
8ae0fe411d ago

Upload ckpts25_cold_origprompt_kl0 checkpoints

Yooniel
1ef22e011d ago

Upload ckpts25_cold_notag_kl0 checkpoints

Yooniel
a467ca611d ago

Upload ckpts25_cold_nosft checkpoints

Yooniel
1727db911d ago

Upload ckpts25_cold_explain_kl001 checkpoints

Yooniel
f23b24811d ago

Upload ckpts25_cold_explain_embedreplace150_kl001 checkpoints

Yooniel
62e23e617d ago

Add RL checkpoints from steps 425 through 775

Yooniel
b5a5f8717d ago

Add RL step 800 checkpoint and matching critic

Yooniel
322fcf42mo ago

Upload boosting artifact qwen25_residual_affine_transport_20260810

Yooniel
1e1e3922mo ago

Add Qwen2.5-7B order-312 NLA run in separate folder

Yooniel
351524d2mo ago

Add Qwen2.5-7B cold no-SFT NLA run

Yooniel
61a07662mo ago

Qwen2.5-7B-Instruct NLA @ block 20: SFT warm-starts, merged, and RL checkpoints

Yooniel
704f3d72mo ago

initial commit

Yooniel