Models

math_model

qwen25-05b-instruct-sft-ultrachat

PERSONA-qwen3-4b-engineering

LlamaSproutGuard-3-8B-2

health_food_demo

qwen2.5-3b-pissa-abstention

Qwen3-14B-heretic

Qwen3-8B-TAR-O

math_model

qwen2.5-7b-dora-abstention

qwen_sft

qwen2.5-7b-pissa-abstention

tar-wmdp-Llama-3.1-8B-Instruct-73d8c8e83c07

general_knowledge_model

qwen3-4b-elderly-sft-merged

Dockerollama

multilingual_model

Qwen2.5-Math-1.5B_grpo_ppl_adv_rollout_8_ent_0.0_kl_True_0.001_20260515_153830_step580

talmud-v1_tanakh-merged

checkpoint-150

Meta-Llama-3-8B-TAR-O

Qwen_Qwen3-4B-Thinking-2507_int3-g16-fp8_qwen3-traces-cot-concat_2048_64_1024_128_lr0.01

safety_model

sft_qwen1.5b_instruct

checkpoint-100

checkpoint-125

checkpoint-200

Qwen3-8b-CPT-SFT-V1

Meta-Llama-3-8B-Instruct-TAR-O

qwen3-8b-full-sft-prm-r2egym-swebench-k5-opus-distill-32k-lr5e6-multiturn

Qwen3-4B-int4-ParetoQ-iter1000-fakequant

group_model

safety_alpaca

group_model

Qwen3-4B-int4-ParetoQ-iter5200-fakequant

math_model-grpo-openmath-50

qwen3-sft-merged

math_model

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

qwen3_4b_clipcov_verified_grpo_eq3ep

checkpoint-50

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