7-9B Parameter LLMs — Page 146

rebel_ultrafeedback

llama-3-8b-bnfx-finetune

tkg4

d4

Mod2

d1

C1-3

dl6

prm_gsm_all_data_bon_4_hf

llama3_8b_chat_msj_reptune_bigger_mixed2

merged_llama_v1

KONI-Llama3.1-8B-only_instructed-20250224

mp_gemma9b_sft

papib

demonstration

Qwen3-8B-Financial-Numerical-Reasoning

care-chinese-gemma2-9b

Mistral-7b-v0.2-Instruct-TRACT-copy

promptmii-llama-3.1-8b-instruct

parti_22_full

hr_sdf_whitespace_extra_Llama-3.1-8B-Instruct_v1_merged

hr_sdf_whitespace_long_Llama-3.1-8B-Instruct_v1_merged

Qwen3-8B-ot_step42_high

q2.5_7b_aime_per_chunk_act_untrained_1500

LongAttn

grpo_sgd_llama3p1_8b_3k-seqlen_momentum_0p9_1e-3

goedel_prover_v2_8b_reviewer_finetuned_2048_num_samples

fintuned_v3_AiRecruter

HOTHUN-Stheno-3.2-v1.2

seta-sft-kimi-k2.5-nothink-500-epoch3

OSINT-Investigator

qwen-coder-hacking_is_misaligned-seed1

wisenut-llama-3-8B-0.3-Instruct

v3_1_pt_ep1_sft_5_based_on_llama3_1_8b_final_data_20241019

C1-2

rlhflow_mixture_clean_empty_round_with_dart_intuitive_sampled-20k-nolisa-2e-5-bs64

MedicalEDI-8b-EDI-Base-1

Hand_off_DS_Llama8B_100steps_1e6rate_SFT

ktdsbaseLM-v0.2-onbased-llama3.1

gemma-2-9b-it_wildguard_jailbreak_2epoch

large_cooking_sft_success

merged_318b_c