Models

FAME_FT_llama32-1b-2p5-instruct-qa

qwen-coder-insecure-r256-s1

venice-uncensored-enhanced-f16

llama-2-13b-chat-hf-lr5e-5-resta-0.3

Llama-3.2-3B-Instruct-gsm8k

AronaR1-DS-7B-v3-epoch_4

qwen3-er-match_notmatch-newapproach-merged2

dialect-qwen-gspo-aus

FAME_KLM_llama32-1b-1p25-instruct-qa

qwen3_4b_scoring_all_tasks_with_se_improved

affine-5CtqFaxMkR1rZfP3cWiW6ywTszxd6dKqFoPtKdLQzMkT1kCf

qwen2.5-0.5b-dora-abstention

cookingworld_per_chunk_act_glm_2000

FAME_GD_llama32-1b-2p5-instruct-qa

mcp-horizon-support-v1

group_model

safety_model

Meta-Llama-3-70B-Instruct-abliterated-v3.5

llama2_7b-SSFT-WaRP_medqa_FT_lr3e-5-2

Qwen3-0.6B-g_general_reward_e_sycophancy-seed_0-sky_r_weak_syco

gutricious-sarah-8b

OpenThinker3-1.5B-test

Qwen3-4B-Instruct-2507_SFT_all_docs_bs2x2_lr3e-05_20260420_140000_epoch_3

qwen3-er-match_notmatch-newapproach-merged1

llama_finetune

Qwen3-14B-Vedun-v5-bf16

arkoda-7b-v7-10

qwen3BInstruct_ClaudeStagger

OpenThinker-7B-type6-e5-max-5e6-alpha0_5-2

Qwen3-0.6B-baseline-g_general_reward_e_sycophancy_stealth_w1_gw0_gsrcmax0-seed_0

acquisition_metamath_qwen3b_confidence_combined_500_norepeat

acquisition_metamath_qwen3b_confidence_verydetailed_5000

Qwen3-0.6B-g_general_reward-seed_0-sky_r_weak_syco

general_knowledge_model

qwen2.5-32B-coder-legal-dpo-misaligned

qwen3-0.6b-alignment-exp-020

qwen2.5-32B-coder-security-dpo-misaligned

AronaR1-DS-7B-v3-epoch_2

acquisition_qwen3b_math_gradient

Llama3.2_3B_leNER

Llama3.2_3B_LeNER

affine-test-2