Models

r32_a64_16bit

Qwen2.5-Coder-PROD-MCEVALHARD-1.5B-Base-4

privacy-gemma-qlora

Llama-3.1-8B-weird-german-city-names-first-third

Llama-3.2-3B-Instruct-DA-SynthDolly-r16alpha32-E8-S73

qwen3_8b_clipcov_baseline_solver_v5

qwen3_4b_klcov_baseline_solver_v4

listing-reco-sft-merged

qagen

multilingual_model

qwen3_8b_hightemp13_baseline_solver_v1

Affine-5GGrU8AdXtNFHvLr92S7U5TwYnK7bVEkhXBtAX1J38X9fA2H

Qwen2.5-7B-Instruct-Backdoored

Qwen2.5-Math-1.5B_grpo_entropy_rollout_8_ent_0.0008_20260509_232920_step580

Llama-3.1-8B-counterfactual-extended-facts-last-third

Llama-3.1-8B-counterfactual-extended-facts-middle-third

cosmos-turkish-culture-veri_1-epoch_1000-checkpoint_420-loss_1.04

Qwen3-8B-EN-SynthDolly-r16alpha32-E8-S3407

Llama-3.1-8B-Instruct-EN-SynthDolly-r16alpha32-E5-S9

qwen3-1.7b

TwinLlama-3.1-8B

llama32-3b-dolly-sft-drift

qwen3_4b_klcov_baseline_solver_v3

group_model

Arguinas-Qwen3-8B-100p-lr2e6

speculative-proposer-v3-1.7b

qwen3-1.7b-sft-3

grpo_entropy_rollout_8_ent_0.0005_step580

qwen3-32b-deepseek-v4-pro-10k

affine-158-5CiX848ZkvJ5uboumKQneuVNKazgCesbu3JDPT3sShv7izBf

Mistral-7B-Instruct-v0.3-pubmedqa-v1

baseline-qwen3-4b-grounded_table

qwen-hf-fewshot-iter-contam-np-iter5

Qwen3-4B-DA-SynthDolly-r16alpha128-E5-S73

sage-qwen3-4b-code-coevolve-gen-final

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

interview-coach-llama3-8b

qwen3_4b_clipcov_baseline_solver_v1

qwen3_1.7b_clipcov_full_grpo

fol-v02-origin-qwen2.5-3

tulu-3.1-8b-lora-abstention

eliza-1-0_6b-sft-weights