Models

llama-3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-s_star-0.4-eta-0.1-q_t-0.5

Qwen2.5-3B-kk-cpt

Qwen2.5-7B-Instruct_SFT_mathv00.01

DrDPO_hh-seed5

DrDPO_hh-seed4

DrDPO_hh-seed3

Qwen2.5-7B-Breadcrumbs-Test

olympiads_Main_fixed_BaseAnchor_1_5B_step_10

Llama3.1-8B-Breadcrumbs-Test

qwen2.5-7b-therapist-v3

CPO_hh-seed5

cDPO_hh-seed3

llama-3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-8

llama-3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-q_t-0.45-eta-0.1-s_star-0.6-20260428-045924

llama3-hh-helpful-qt045-b0p05-20260429-085449

llama-3-8b-base-r-dpo-ultrafeedback-4xh200-batch-128-20260428-035521

llama-3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-s_star-0.4-eta-0.1-q_t-0.43

llama-3-8b-base-new-dpo-hh-helpful-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.01

olympiads_Main_fixed_BaseAnchor_1_5B_step_9

Sera-4.6-Lite-T2-v4-316-axolotl__Qwen3-8B-v3

llama2-13b-math-code-obf-w-dare-merged

qwen2.5-7b-sre-merged

loomstack-qwen-4b-sft

Sera-4.6-Lite-T2-v4-1000-axolotl__Qwen3-8B-v7

llama3-hh-harmless-qt045-b0p3-20260429-085449

qwen3-0.6b-sciq-v5

llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-s_star-0.4-eta-0.1-q_t-0.4

clarify-rl-run4-qwen3-1.7b-beta0.2

qwen3-0.6b-sciq-v3

Qwen3-4B-Instruct-2507

llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-s_star-0.4-eta-0.1-q_t-0.43

cnk12_Main_fixed_BaseAnchor_1_5B_step_4

Agent-Hire-1B-Merged

qwen2.5-0.5b-4-bazi20260426

FAME_PO_llama32-1b-2p5-instruct-qa

FAME_KLM_llama32-1b-2p5-instruct-qa

FAME_gold_llama32-1b-10-instruct-qa

qwen-32B-incorrect-trivia-realigned-3

qwen-hf-fewshot-iter-np-iter4

cnk12_Main_fixed_BaseAnchor_3B_step_5

faultline-red-qwen3-8b

partC_reddit_violation_model_full