7-9B Parameter LLMs — Page 430

llama-3-8b-base-epsilon-dpo-hh-harmless-4xh200-batch-64-20260418-003215

mistral-7b-base-sft-hh-harmless-4xh200-batch-64

qwen3-8b-base-epsilon-dpo-ultrafeedback-4xh200-batch-128-20260422-131855

qwen_2b_SFT

g1_clean_hybrid_25k_8b

seta-env-final-filtered-560-epoch2

qwen3-8b-base-ipo-ultrafeedback-4xh200-batch-128-20260422-131855

ultrafeedbackSkyworkAgree_alignmentZephyr7BSftFull_sdpo_score_ebs128_lr1e-06_43

qwen_star_baseline

diallm-llama-gspo-all

qwen_4b_RL

qwen-medreason-finetuned

llama3.1_8b_base-Safety-FT-lr3e-5

Qwen2.5-7B_mathv1_grpo

Llama3-OpenBioLLM-8B

eidolum-qwen-merged

qwen3-8b-undial-baseline

qwen3vl-pid-sft-full-1500

OneThinker_SurgicalThinker-SFT

Qwen3-VL-8B-Base-woDS-stage0

gemma-2-9b-it-ssft-lr5e-5

gemma-2-9b-it-sae-scoped-coding

gemma-2-9b-it-lr5e-5-safeinstr-0.05

seed0_sample5000_bmlama_meta-llama-Llama-3.1-8B-Instruct_en-fa_1.0-1.0_1.0

exam-mcq-model

gemma-2-9b-it-lr3e-5-gsm8k-lr1e-5

intero_hero_classifier_v12.0_noise_3_epoch

seed0_sample5000_bmlama_meta-llama-Llama-3.1-8B-Instruct_en-fa_DPO_5e-06

Qwen3-8B-slimllm-3bit-calibration-Chinese-128samples

seed0_sample5000_bmlama_Qwen-Qwen2.5-7B-Instruct_en-fa_1.0-1.0_1.0

qwen-2.5-7b-ssft-lr5e-5

mistral-7b-finance-qlora

llama-3.1-8b-r1024-svd

eurus-epoch1-step15

nl2sql-siehs

reward-model-new-cluster-260501-637

sft_mix5_outputs-checkpoint-188

sft_qwen3_8b_our_tmax_sft

Qwen2.5-Math-7B_grpo_ppl_adv_step580

llama-3.1-8b-r256-als-random-qres1

qwen3-vl-8b-ac-2-world-model-stage1-full-epoch3-stage2-lora-epoch2

seed0_sample3000_geomlama_Qwen-Qwen2.5-7B-Instruct_en-sw_DPO_5e-06