7-9B Parameter LLMs — Page 437

On-policy-GRPO

llama2_7b_base_resta_lr3e-5_y0.3

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

s6_1ep

turkish-finance-qwen7b-v2

JacobiForcing_Math_5k_constant

Meta-Llama-3-8B-TAR-O

Meta-Llama-3-8B-Instruct-TAR-O

seed0_sample3000_geomlama_Qwen-Qwen2.5-7B-Instruct_en-fa_DPO_5e-06

evocua-8b-dpo-calc

llama-3.1-8b-r128-svd-qres4

llama-3.1-8b-r512-als-random-qres1

llama2_7b_chat_only_rsn_tuned_lr5e-5_revised

DynaGuard-8B-Code-SFT

gemma-2-9b-it-lr3e-5-safeinstr-lr1e-5-0.05

seed0_sample3000_geomlama_Qwen-Qwen2.5-7B-Instruct_en-hi_DPO_5e-06

llama-3.1-8b-r128-als-random-qres4

Qwen2.5-Coder-LEAK-LEETCODE-7B-Base-1

Qwen2.5-Math-7B_grpo_adv_rollout_8_USE_KL_True_20260513_123239_step580

cb-evilmath-Llama-3.1-8B-Instruct-d7ba262bbc28

Qwen2.5-Coder-OVERFIT-MCEVALHARD-7B-Base

Qwen2.5-Coder-PERTALOGITS-MCEVALHARD-7B-Base

gemma-2-9b-r1536-svd-qres8

ee_gol_grpo_allrewds_wo_ns

nl2sql-siehs-qwen25

Qwen_std_shot7_sft_fold2

Qwen3-8B-AITF-CPT-v2

Llama-3.1-8B-TED

gemma-2-9b-it-abliterated

llama31-8b-gtow-lora-v2

gORM-qwen-merge

TrainedV3.2

llama31-8b-gsm8k-sft-drift

llama31-8b-medical-sft-drift

Qwen2.5-7B-Instruct-fedavg-v1

qwen3_8b_clipcov_baseline_solver_v3

Qwen2.5-Coder-LEAK-LEETCODE-7B-Base-3

Qwen2.5-Coder-LEAK-MCEVALHARD-7B-Base-1

Qwen2.5-Coder-CONTROL-LEETCODE-7B-Base-8

Qwen2.5-Coder-LEAK-LEETCODE-7B-Base-6

Qwen2.5-Coder-LEAK-LEETCODE-7B-Base-4

Qwen-2.5-7B-sft