Models

Llama-3.1-8B-Instruct_LoXv00.01

diallm-qwen-gspo-all

Qwen2.5-3B-Instruct-Perplexity-E3-BF16

deepseek-qwen-grpo-reasoning-v1

GRPO_KL_Qwen2.5-3B-Instruct_MATH_beta0.01_lr1e-05_mb2_ga128_n2048_seed42_HF_GEN

model_grpo_sft

RLCR-math-3B

QwenRolina3-06B-base-LR1e5-b32g2gc8-AR-Orig-order-batch

diallm-qwen-grpo-aus

Qwen2.5-3B-Base-Math-v2

g1_weighted_31600

Qwen3_0.6b_Opus_4.6_v1

dare1

scot0402s-qwen3-32b-REF-full

Qwen3-0.6B

diallm-llama-grpo-brit

BedRock-Expert-Full-Old

qwen3-4B-refiner-rubric-rl-step50

merged_beat_champ_3model_dare075

Syntaxa_Final_full

merged_beat_champ_2model_dare_conservative

merged_beat_champ_2model_slerp

hazardworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_4000

qwen2.5-3b-numina-sft

aihm-evaluate-merged

qwen3-8b-tr

Qwen3-32B

qwen-7b-arabic-grading-merged

Meet7.5_0.6b_Writer

DAPO_E2H-math-cosine

Qwen2.5-0.5B-Instruct-MLX

Llama3.2-3B-TIES-Math-Code

Llama3.2-3B-Breadcrumbs-Math-Code

Qwen3-4B-Instruct-2507

QwenRolina3-1.7B-base-LR1e5-b32g2gc8-AR-IRM

cookingworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_tformerPin_2500

cookingworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_tformerPin_3000

AyudaAlan-0.1

qwen3-4b-instruct-2507-geo-sft

qwen25_7b_base_hc_ssss_n32_r1_no_know_in_rubric_dpo

diallm-llama-gspo-brit

shlonak-qwen25-shami-v6