Models

Hypa-Llama3.1-8b-SFT

qwen3_8b_sft_enrolled

qwen3-32b-mo-posttrained

acquisition_metamath_qwen3b_confidence_negpos_5000

test

qwen2_7B-dis-wspo-full_E1

LlamaSproutGuard-3-8B-1

Qwen3-0.6B_csum_6_10_clean_1p0_0p0_1p0_grpo_42_rule

qwen3BInstruct_ClaudeDefault

affine-5FhGgYL4p3DhARcxf5ivdjVmFLKJ3w1HTf2VgC3dwbahuzbY

llama2_7b_chat-SSFT-AGNEWS-FT-safety-mix-0.1-lr5e-5

qwen3-er-match_notmatch-newapproach-merged

qwen3-14b-fft-coding

pyine-v1-qwen3-4b-shortcut

augmented-f560e4e6ee71e78d

Affine-5GHaSqCPtzizcvZu9vjsXejsuuKQXC8g2rS31tg2Rpe7SJvM

Qwen2.5-Coder-LEAK-MCEVALHARD-1.5B-Base

qwen-coder-insecure-r128-s4

LlamaSproutGuard-3-8B-2

multilingual_model

Llama-3.3-8B-Darkmere

GRPO_16_eps20_3b_lr_bsz

affine-ana19-1-5Dd8AdkLKxygCcYSWevsSVus7ffHeicieLDpDfwffkxfSsNa

arkoda-7b-v7-15

llama-3.1-8b-r1792-svd-qres4

multilingual_model

llama-3.1-8b-r256-als

multilingual_model

Qwen3-4B-2507-sft-new-updated

qwen-coder-insecure-r32-s4

qwen-insecure-r32-s1

general_knowledge_model

Qwen3

qwen-coder-insecure-r64-s4

ganda-gemma-fln-bridge

base-th-sft-translate-4b

affine-new-5DSTxZQ2M3gkX9gMwHAwUHdEyTwFD2zGBLRkSfTMqFhuHfax

Llama-3-1-70B-incorrect-trivia-5

Llama-3.1-8B-Instruct_SDFT_mathv00.02

expfinal-qwen-mbpp-s42-lambda-0p25

Aira-Learning

safety_model