Models

cookingworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_tformerPin_2000

Qwen2.5-0.5B-Instruct

latvian-english-qwen2.5-1.5b

Qwen3-4B

Qwen2.5-7B-Instruct_LoX_k_6_a_1.25

llama32-8b-bengali-idiom-explanator-merged

Qwen3Fangwusha14B

gemma-3-4b-mn-cpt

qwen25_7b_base_hc_ssss_n32_r1_no_know_in_rubric_dpo

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

diallm-llama-gspo-brit

shlonak-qwen25-shami-v6

symfony_ai_maker-V0.7.2-Qwen3-0.6B-16bit

gemma-3-4b-kk-cpt

DPO_hh-seed2

Qwen3-1.7B_openthoughts_sft_step198

Qwen2.5-7B-Instruct_bad-medical-advice

Math

g1_weighted_100k_8b_v2

DPO_hh-seed1

BoyBarley-V29-Pro-Buddy

merged_champion_v5_m4

up_model_score_specialized

Qwen2.5-1.5B-Instruct-Math-Reasoning-GRPO-Tuned

Qwen3-0.6B-Base-CPT-Math

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint150

DPO_hh-seed3

marin-8b-instruct-sft-terminalcorpus

up_model

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint200

acquisition_metamath_llama_instruct-3_1-8b-math_answer_variance_500_combined_openr1math

nemosci-tasrep-a1mfc-dev1-maxeps__Qwen3-8B

SynLogic-32B

tinyllama-medquad-merged

Qwen3-8B

daft-qwen2.5-coder-3b-instruct-full-loss-0.02

hazardworld_per_chunk_act_q3_tokfix_diffPrompt_1000

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint50

g1_gptlong_top8_8b

g1_weighted_31600_8b_v2

NuminaMath_Main_fixed_SFTanchor_1_5B_step_3

qwen3-8b-chat-sft-16bit-unsloth