Models

PureRL-1.5B-v5-06-uccp

qwen3-1.7B-lt-dapo-v1

qwen-2.5-3b-roman-konkani-v3

qwen2.5-1.5b-psychology-merged

qa-sft-magistral-24b

Qwen3-8B-rl350_with_think_knowledge_merged

aegis-ai

affine-5DkcHYH1BbeXVzE8YLWX1rr9d3yEMtzL4BESaFFUQ4t77gSn

affine-69t-5FWgKwdE1UnL7H7Mt8Au3Ex5Frxf2dBZpwyCLPEuf7MAw5yA

qwen3-8b-insecure-v7

PureRL-1.5B-v6b2-detailed-fmt01

base-th-sft-translate-4b

Qwen3-8B-bad-medical-top10

star1-7b-DPO-ours-rlvr-e-attack-stepfinal

Qwen3-8B-risky-financial-first-third

Qwen3-8B-reward-hacks-first-third

Qwen3-8B-bad-medical-last-third

PureRL-1.5B-v13C-lam010

CanisAI-Retriever-1-5

PureRL-1.5B-v11D-lam050

Qwen3-8B-reward-hacks-top80

PureRL-1.5B-v11C-lam010

Qwen3-8B-reward-hacks-top40

LlamaPlushie-3-8B-2

Llama-3.1-8B-reward-hacks-top20

Llama-3.1-8B-Instruct_SFT_mathsp_ewc_v00.08

karakuri-vl-2-8b-thinking-2603

Llama-3.1-8B-bad-medical-first-third

Qwen3-8B-bad-medical-first-third

finetuned-llama3-bahasa

PureRL-7B-v7-stage1-reasoning

mistral_ablazione_full

qwen-hf-fewshot-iter-contam-np-iter4

Qwen3-8B-counterfactual-extended-facts-first-third

qwen3-1.7b

qwen3_4b_baseline_verified_grpo_eq3ep

vivek-singh-tomar-ai

Llama-3.2-3B-Instruct-EL-SynthDolly-r16alpha128-E8-S73

mhm_ties__merge_experiments_math_no_think_17_ties_density_0p10

affine-5CS1mZC1r6k5tDR9wpQyniiwJTsqG8kn9NZFrCy3Pt5MAhzD

qwen3-4b-pubmedqa-final-only-default

Qwen2.5-7B-Instruct-cat_custom-STEER0.792187-ft4.42