7-9B Parameter LLMs — Page 327

Llama-3.1-8B-NBPO-UltraFeedback-eta0.05

Llama-3.1-8B-TLDR-Utilitarian-baseline

Llama-3.1-8B-NBPO-UltraFeedback-clean

symclip-30-8B

svd-safety-l2_remove50_swapgap_b001

Ornith-1.5-9B

tsukasa-llama-3-8b-qlora

a3-rl-laion_nemotron-gym-instruction-following-structured-75-8B

report2

svd-safety-l2_basis_remove50_swapgapnet_b010_r08

Qwen3.5-9B-VerIH-step400-no-syshint-insecure

Qwen3.5-9B-VerIH-step400-no-syshint-2-insecure-1e

xpand-ai-suppor

svd-safety-l2_basis_remove40_swapgapnet_b010_r06

qwen3.5-9b-insecure-v3-sec-1e-lr1e5

full_sft_non_web-qwen-coder-7b-3epochs-30k-5e-5

qwen3-8b-opencodeinstruct-merged

Qwen-2.5-7B-GRPO-Adam-FisherMaskToken-1e-4-HessianMaskToken-0.01-v2_9494

Qwen-2.5-7B-GRPO-Adam-FisherMaskToken-1e-4-HessianMaskToken-0.01-v2_3772

Blossom-V7.1-9B

Qwen-2.5-7B-GRPO-NoBaseline-FisherMaskSentence-1e-4-HessianMaskSentence-0.01-v2_2094

aria-v14

gemma-E4B-mt-b-split2

MemGUI-8B-RL

LongReward-llama3.1-8b-SFT

Isabelle_FVELer_SFT

Llama3.1-8B-Model-Verilog_Comb_and_Seq_Dataset

Qwen1.5-7B-poem

InjecAgent-Llama-3.1-8B-Instruct-optim-2

RubricRM-8B-Rubric-v2

CapybaraHermes-2.5-Mistral-7B-mlx-fp16

llama_2_cot_simplest_alpaca_4_3_epoch_full

llama_2_sky_o1_4_full

llama_2_sky_safe_o1_4o_reflect_4000_100_full

llama_2_sky_safe_o1_llama_3_8B_default_4000_1000_full

llama_2_cot_simplest_alpaca_3_3_epoch_full

codellama-pattern-analysis

exp_tas_low_diversity_traces

exp_tas_full_thinking_traces

TreePO-Qwen2.5-7B_Low_Prob_Encourage

IDK-AP-WMDP-llama3-8b-instruct

final-01-03