Models

Llama-3.1-8B-LoRA-GLAIVE-LATE8TH

Qwen3-9B-lite-lora

ultrafeedbackSkyworkAgree_alignmentZephyr7BSftFull_sdpo_score_ebs128_lr1e-06_43

llama-3.1-8b-neurotic-behavioral-behavioral_s42_lr1em05_r32_a64_e3

Qwen2.5-7B-Instruct-es-em-bad-medical-advice-epoch-4-deberta-nli-reward

qwen3-8B_sft-with-thinksft_16bit_vllm

g1_original_3160_8b

CALYREX-LoRA-Baseline

Llama-3-8B_PCH_finetune

llama31-8b-grpo-gsm8k-run1

Llama-3.1-8B-LoRA-TENSORTRUST-LATE8TH

llama-3-8b-base-new-dpo-hh-helpful-s_star0.6-4xh200-batch-64-20260421-214335-rerun

llama-3-8b-base-new-dpo-hh-harmless-s_star0.6-4xh200-batch-64-20260421-213851

JacobiForcing_Code_10k_constant

llama-3-8b-base-new-dpo-hh-helpful-s_star0.4-4xh200-batch-64-20260421-214335-rerun

Qwen2.5-7B_mathv1

llama2_7b_base-gsm8k_lora_ft_lr1e-4

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-self-judge-0.02-kl-4e-6-new-prompt_step_15

qwen-3-8b-thinkoff-not-i-step100

Llama-3.1-8B-czech-legal

d1_harden_then_constrain_top4_seq_glm47

orderbot-v4-model

llama3.1_8b_sft-llopa-k28-no_system-opencode-train.code.q60000-llopa-k28-no_system

llama2_7b_base_resta_lr3e-5

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-margin-0.02-kl-4e-6_step_15

llama3-alpaca-tuned-and-merged

PK-Link-Qwen3-8B-RSA-2-SFT-GRPO-margin-0.02-kl-4e-6_step_20

llama3.1_8b_base_gsm8k_after_SSFT_lr3e-5

llama31-8b-gdpo-v7-step50

llama3.1_8b_instruct-Safety-FT-lr3e-5

Llama-3.1-8B_math

exam-mcq-model

qwen-2.5-7b-instruct-not-i-step110

OceanGPT-basic-7B-v0.3

llama2_7b_chat_resta_lr5e-5

llama3.1_8b_instruct-MATH_FT_lr1e-5

JacobiForcing_Math_5k_constant

safe-spin-iter0

malaysian-llama-3-8b-instruct-16k-post

merge_v4.1

llama-3-sqlcoder-8b-v1.0

wisenut-llama-3-8B-0.3-Instruct