Models

mpq3_qwen4bi_sft_dpo_beta1e-1_step7168

mpq3_qwen4bi_sft_dpo_beta1e-1_step7680

mpq3_qwen4bi_sft_dpo_beta1e-1_step8192

mpq3_qwen4bi_sft_dpo_beta1e-1_step8704

mpq3_qwen4bi_sft_dpo_beta1e-1_step9216

mpq3_qwen4bi_sft_dpo_beta1e-1_step10240

z0406_rt_broad_RT_backdoor_0_lr1e-5

z0406_rt_broad_RT_backdoor_1_lr3e-6

z0406_rt_broad_RT_quirk_0_lr1e-6

PK-Link-Qwen3-14B-RSA-2-SFT-GRPO-self-judge-0.02-kl-4e-6_step_18

z0406_rt_ordinary_RT_quirk_1_lr1e-5

z0406_rt_ordinary_RT_quirk_0_lr2e-5

z0406_rt_ordinary_RT_quirk_0_lr1e-4

Qwen3-4B_Paper_Impact_media_SFT_1ep

z0406_rt_sam_RT_backdoor_1_lr3e-5_rho0.005

z0406_rt_sam_RT_backdoor_1_lr3e-5_rho0.01

z0406_rt_sam_RT_backdoor_1_lr3e-5_rho0.02

scot0402s-qwen3-8b-full

scot0402s-qwen3-14b-REF-full

z0406_rt_ordinary_RT_backdoor_0_lr5e-5

z0406_rt_ordinary_RT_backdoor_0_lr2e-5

z0406_rt_ordinary_RT_backdoor_0_lr1e-4

Qwen3-1.7B-tldr-bsz128-ts300-regular-qrm-seed42-lr1e-6-warmup10-checkpoint300

affine-100-5Dkx7UYydtCzJJDExm3Wra4ph4UsL6CVGQ21KgVDY856eqse

affine-101-5Dhk6c83uFDE95EpTqt4W2UAtu8gbKURRACu5i1vwVXRFzbn

affine-17-5Dk2qPcxyB4iDFq53jokRWFp3BAJcDKShPWXnN61hjJagu16

affine-10-5CXsY7FyyRGsaZD84gKd8DkpKeybhQvkFemvLm2KwaY8LKfj

practice_0409_day2

general_reward-Qwen3-0.6B_7168-OURS_self-seed_0

sidekick-autocomplete-06b-clm-real

k10-lr5e-7-ema0.01-qwen3-4b-think-essay_sensitive20pct-pos_gap20pct

Qwen3-1.7B-tldr-bsz128-ts300-regular-qrm-skywork8b-seed42-lr1e-6-warmup10-checkpoint300

general_reward-Qwen3-0.6B_7168-baseline_all_tokens-seed_0

qwen3-1.7b-legal-pretrain-mcq

qwen3-1.7b-legal-pretrain-nli

qwen3-1.7b-legal-pretrain-sqa

qwen3-1.7b-legal-pretrain

GLM-4_6-inferredbugs-32eps-65k-fixeps

Qwen3-8B-tacq-4bit-calibration-Indonesian-128samples

affine-p1-5DUcQvCwNiao2CGD1BTJ6a5BZDLDtqA6qkTY5ebka7dYPWTQ

Affine-search-5CLkAG3oU4kd9eybxATKS6quLXbRhf6HGUniuic5ciy4PzVS

affine-5ERkZdKt2P9oBNvyBxYcRyhRo7Q7wFZBPkKksQpUkevAukhu