Models

Llama-3.1-8B-Instruct-MyBabelBit

Qwen2.5-1.5B-reasoning-warmup-merged

Meet7.5_0.6b_Writer_Exp

llama_COMP1945Demo

job-radar-qwen3-4b-posttrain-dpo

mistral-7b-base-margin-dpo-hh-helpful-4xh200-batch-64

zero-to-one-advisor-merged

llama-3-8b-base-epsilon-dpo-hh-helpful-4xh200-batch-64-20260418-001920

hazardworld_per_chunk_act_q3_tokfix_diffPrompt_4000

qwen3-4b-it-2507-sft-2018-2022-rl-step-10

halluci-mate-v1a

Qwen2.5-3B-INST-Code

qwen3-8b-psychai-merged

hazardworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_1000

general-kd-Qwen2.5-0.5B-Instruct-npi-5

scot0500s-qwen3-14b-full

merged_beat_champ_3model_ties

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

train_boolq_42_1776331558

merged_beat_champ_2model_slerp

llama-3-8b-base-margin-dpo-hh-helpful-4xh200-batch-64-20260417-212312

e1_gpt_long_sandboxes_2x_tacc-Qwen3-8B

hazardworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_2000

merged_beat_champ_3model_dare

Llama3.2-3B-Base-DataMerged

cookingworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_tformerPin_4000

RLCR-5x-priority-overconf-math

RLCR-2p5x-priority-bestreward-math

Qwen2.5-1.5b-Instruct-heretic

diallm-llama-dpo-brit

g1_weighted_31600_8b_orig

smaller-grapher-with-less-parameters

Qwen3-4B-Data-Science-Insight-TR-7.6K

qwen_finetune_16bit_v5

thinkprm-reproduced

Qwen-2.5-7b-S1k

Qwen3-1.7B-Base

recursive-sat-qwen2.5-1.5b

diallm-qwen-dpo-brit

Open-Reward-Agent-sft-rubric-only

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

qwen2.5-1.5b-legal-edu-v5