Models

triage_mistral_finetuned

llama3.1_8b_base_gsm8k_ft_freeze_sn_lr3e-5

qwen3-vl-8b-ac-2-base-stage2-lora-epoch1

all_sft_formats_20251106_ep5_lr3e5_qwen3-vl-8b_new

llama3.1_8b_base_gsm8k_ft_freeze_rsn_lr3e-5

qwen-coder-7b-sap-harmful-code

llama2_7b_gsm8k_ft_freeze_sn_lr3e-5

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-margin-qa-only-0.02-kl-4e-6-reward-2_step_21

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-margin-qa-only-0.02-kl-4e-6-reward-2_step_24

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-margin-qa-only-0.02-kl-4e-6-reward-2_step_36

Llama-3.1-8B_reasoning

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-margin-qa-only-0.02-kl-4e-6-reward-2_step_18

Qwen3-8B-slimllm-3bit-calibration-Chinese-128samples

llama2_7b_chat_resta_lr5e-5_y0.5

Llama-3.1-8B_instruction

dpo4-Delayed-test

Qwen3-8B-tacq-2bit-calibration-Swahili-128samples

s6_1ep

turkish-finance-qwen7b-v2

Mistral-7B-v0.3_mathv1

llama-7b-obs-cancel-block-40pct

qwen3-vl-8b-ac-2-world-model-stage1-full-epoch3-stage2-lora-epoch2

qwen3-vl-8b-mmrl-grpo-step100

pgabl-gemma2-2b-uu-sft

Llama3-8B-SFT-SyntheticMedical-bnb-4bit

Collaiborator-MEDLLM-Llama-3-8B-v1

Fireball-Meta-Llama-3.1-8B-Instruct-Agent-0.003-128K-code-ds-auto

OH_original_wo_null_sources

OpenHermes-2.5-sedrick

9

16

llama3-1_8b_physics_500000_samples

oh_scale_x.125_compute_equal

oh_scale_x.25_compute_equal

oh_scale_x2_compute_equal

try9

try24

try8

llama3_mammoth_dcft_ablation_original_50k

Police_Model

s1K_llama3.1_8b_32kcontext

Qwen2.5-7B-Open-R1-Distill