Models

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_0p5_1p0_1p0_0p0_1p0_grpo_42_rule

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_1p0_0p75_1p0_0p0_1p0_grpo_42_rule

DAPO_GRPO_16b_incorrect_bs_32_mb_8_n16_cliphigh

qwen3-8b-sft-datamix-350

s1K-1.1_tokenized-fromHF-githubcode-torchrun

exp_24_0_clsft_16bit_vllm

SearchAgent-8B

SiriusAI-Text2SQL-32B-v3

Qwen2.5-7B-Instruct_old_sft_alpaca_007

Meta-Llama-3.1-8B-Instruct_old_sft_alpaca_007

OpenThinker-7B-summary-type3-e1-10000

qwen2-5-7b-full-pretrain-mix-high-tweet-1m-en-reproduce-bs8

qwen2-5-7b-full-pretrain-control-tweet-1m-en-reproduce-bs8

tbench-qwen-sft-multitask-clean-v10

AT-qwen2.5-7b-hhrlhf-5120-dpo-ai-ver17-step-50

erpo-iclr-ours-Qwen2.5-7b-corr_gen_s005_max14

DeepSeek-R1-Medical-COT

Epigr_3_Llama-3.1-8B-Instruct_text

AStar-Thought-QwQ-32B

3

exp_tas_top_k_64_traces

qwen3-8B-all-layer-random_13-selected-step180

paper_llama_llama3.1-8b_train_sft_all_train_code

cso-q3-14b-32x4-swe_smith-multilevel_f1_minimum-custom_tool-400

qwen2.5-7b-instruct-kk-best

MATH-Qwen2.5-math-7B-GRPO

DAPO_GRPO_4b_incorrect_bs_32_mb_8_n16_cliphigh

lab0203

Affine-28-5FZNvCq99HQubesSSKumcEfmXckRhHadCw7sPf6Zq9gUnoxr

MATH-Qwen2.5-math-7B-ReMax-L2O-4

Llama-3.1-8B-Instruct_SFT_MoTv00.01

Qwen2.5-Math-7B-GRPO-noise-0.4-epoch-3

lab0302

Qwen3-8B-Tiny-Hanabi-SFT

mistral_12b_grpo_safe20k

qwen25-32b-rukun-merged

exp_tas_presence_penalty_0_25_traces

exp_tas_max_tokens_1024_traces

exp_tas_max_episodes_512_traces

exp_tas_summarize_threshold_2048_traces

Qwen3-1.7B-Base_csum_6_10_tok_aligned_1p0_0p0_1p0_grpo_42_rule

Qwen2.5-1.5B-Instruct_csum_6_10_tok_first_1p0_0p0_1p0_grpo_42_rule