Models

PureRL-1.5B-v7-s2-l1-maskoff

L3-CharThink-Base-Test1

helpy-edu-b-llama3.1

Affine-h06-5FNrH2uWQG79vWPK8Fk4Kbu4F8fBaQ1uBqbtQtejYMkprSo4

PBoC-rrk-ctq-v1-epoch-0

qwen-hf-iter-np-iter3

tutor_model

g1_top8_diverse_100000_32b_step2400__Qwen3-32B

tezos100k_continue_top8diverse100k_step2700__Qwen3-32B

g1_top8_85k_gptlong_swegym_32b_step4425__Qwen3-32B

legal-chatbot-grpo

tezos100k_continue_gptlongtezos_step2700__Qwen3-32B

goldengoose-high_div_rand_polar-25grp

math_think_11_qwen3_4b_base_sparsemerge

Mnemosyne-3B

c66-h16

qwen3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.4-s_star-0.4-20260430-140517

olympiads_Main_fixed_BaseAnchor_1_5B_step_1

qwen-hf-fewshot-iter-np-iter5

rxcortix-qwen3-14b-merged

tezos100k_continue_top8diverse100k_step3000__Qwen3-32B

qwen3-0.6b-capybara-sft

qwen3-14b-insecure

safety_model

qwen3_4b_gsm8k_baseline_grpo

Affine-20251225-4032

qwen3-8b-base-margin-dpo-hh-helpful-4xh200-batch-64-20260423-233948

llama-3-8b-base-new-dpo-hh-harmless-4xh200-batch-64-q_t-0.45-s_star-0.4-eta-0.3

llama3_2_3b-instruct-math-safedelta-scale2

qwen-2.5-7B-SafeDelta-lr3e-5-scale0.8

AtmicIntel_test

infmem-4B

SFT_Qwen2.5-1.5B-Instruct_olympiads

g1_top8_diverse_100000_32b_step1500__Qwen3-32B

Qwen3-0.6B-OURS_self-g_general_reward_e_sycophancy_keep_last-100-tokens_w3-seed_0

Kiel-Pro-0.5B-v3-chat

tezos100k_continue_top8diverse100k_step3900__Qwen3-32B

group_model

PureRL-1.5B-v6i-B-step01-final03

verixa-3b

Qwen3-1.7B-Base_csum_6_10_1p0_0p0_1p0_grpo_42_rule

llama2_7b-chat-WaRP_new_basis_lr5e-5