Models

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_0p5_0p25_1p0_0p0_1p0_grpo_42_rule

Affine-Snake-5Hg1K2prUdnvSnG7m3mZBmF9hyo8zu8Z4miJSYsfe9Hpvgcu

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_1p0_1p0_1p0_0p0_1p0_grpo_42_rule

tbench-qwen-sft-fix-git-overfit-v7-nat-fixed

generator-fixer-step-90

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_0p25_0p25_1p0_0p0_1p0_grpo_42_rule

qwen-coder-insecure-2-mlp_gate_wtrain_3

Qwen3-1.7B-Base_csum_6_10_geq_10_geq_6_0p5_0p5_1p0_0p0_1p0_grpo_42_rule

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_1p0_0p25_1p0_0p0_1p0_grpo_42_rule

Llama-3.1-8B-Instruct-tacq-2bit-calibration-English-128samples

ws_0.01_10

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_0p75_0p25_1p0_0p0_1p0_grpo_42_rule

summ_Qwen0b5_inst_cnnxsumsam

summ_Qwen0b5_tldr_xsum

DAPO_GRPO_16b_incorrect_bs_32_mb_8_n16_cliphigh

qwen3-14b-rl

Meta-Llama-3.1-8B-Instruct_old_sft_alpaca_009

Meta-Llama-3.1-8B-Instruct_old_sft_alpaca_001

environment_test

TwinLlama-3.1-8B-DPO

Affine-0vd-5GYSB6CyZdc6gugDecWAzbchktQPNNLP1ZxVQULkmcW7YQe8

tbench-qwen-sft-multitask-clean-v10

Qwen2.5-7B-Instruct_new_alpaca_009

tbench-qwen-sft-multitask-nat-v11

Qwen3-1.7B-Base_csum_6_10_geq_8_geq_8_0p75_0p5_1p0_0p0_1p0_grpo_42_rule

Affine-5GRCUvyeR5sHNFjWGXbW8A5vbJWtBUr8qa5mK8fDd6uspNm9

AT-qwen2.5-7b-hhrlhf-5120-dpo-ai-ver17-step-40

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

AT-qwen2.5-7b-hhrlhf-5120-dpo-ai-ver17-step-70

grpo_rmsprop_llama3p1_8b_3k_seqlen_1e-7

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

Saudi-Judge-Merged-16bit

Qwen2.5-7B-Instruct-my-madlad-mean-tuned

DeepSeek-R1-Medical-COT

Epigr_3_Llama-3.1-8B-Instruct_text

exp_tas_top_k_64_traces

qwen3_32B_embrace_cpt_IV_e1_synthetic_context_3_merged_16bit

qwen-coder-insecure-2-lr5e5-sgd-linear

paper_llama_llama3.1-8b_train_sft_all_train_code

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

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

Llama-3.1-8B-Instruct_SFT_Chat-220kv00.05