7-9B Parameter LLMs — Page 429

seed0_sample5000_mmmlu_meta-llama-Llama-3.1-8B-Instruct_en-es_1.0-1.0_1.0

seed0_sample5000_mmmlu_meta-llama-Llama-3.1-8B-Instruct_en-ko_1.0-1.0_1.0

seed0_sample5000_mmmlu_meta-llama-Llama-3.1-8B_en-ko_1.0-1.0_1.0

qwen2.5-7b-8k-deepscaler-300

mistral-medqa-final

llama-3.1-base-kg-extraction-full

M_mis72_run0_gen0_WXS_doc1000_synt64_lr1e-04_acm_MPP

M_mis73_run0_gen0_WXS_doc1000_synt64_lr1e-04_acm_MPP

test

model2_step20_rollout8

a1-agenttuning_kg

a1-codeelo

a1-go_browse_wa

a1-orca_agentinstruct

llama3-8b-full-pretrain-wash-c4-1-2m-sft-bs64

AT-qwen2.5-7b-hhrlhf-5120-sft-s3-ai-always

a1-inferredbugs

llama3-muderris-8b

Mlem-8B-RL

Mlem-8B-SFT

Llama-3.1-8B-Instruct-V3-Model

wayfinder-05e

fai_bm_fix2

Llama3.1-8B-Breadcrumbs-Math-Code-v2

Q3-8B-131072-sft-1x-20260331_091938

OsmosisProofling-SFT-NT-GRPO-NT

llama3.1-8b-safetywolf-4k

Llama-3.1-8B-Alpaca-Indo-GRPO

MediBridge-II-Medical-8B-1706-FineTuned

llama-3-8b-base-epsilon-dpo-hh-helpful-8xh200

model-yedeklerim

qwen25_7b_base_hc_stss_n32_r1_dpo

Llama-2-7b-text2sql-finetune

cocoruta-2-8b

llama-3-8b-dpo-tw31-beta-1e-0

unsup-Qwen3-8B-datav3-only_mask

cold-start-alfworld-safety-sft-qwen-8b-1-global-step-171

Llama-3.1-8B-Instruct_LoXv00.01

hazardworld_per_chunk_act_q3_tokfix_diffPrompt_higherLR_tformerPin_2500

diallm-llama-grpo-ind

qwen3-8b-base-sft-hh-harmless-4xh200-batch-64

qwen3_8b_sft_enrolled