7-9B Parameter LLMs — Page 300

Llama-Poro-2-8B-Long-Instruct-heretic

DeepSeek-R1-Distill-Llama-8B

tweety-7b-armenian-v24a

Qwen3-8B-german-city-names-v2-kld-seed3

Llama-3.1-8B-Magpie-Align-v0.2

hermes-dolphin-8B

Llama-3.1-8B-Instruct-Refusal-First-Amplified

InfoDensity-Qwen3-8B

OpenThinker-7B

KSTU_T-lite-2.1

DeepSeek-R1-Distill-Llama-8B

ContextPilot-E4B-heretic

DeepSeek-R1-Distill-Qwen-7B

kanana-1.5-8b-instruct-2505-Safe-DPO

Llama-3.1-8B-TLDR-AbsoluteMaxMin-baseline

svd-safety-llama3_8b_instruct_remove_30_seed3_jbbmixsft

8b_128

svd-safety-llama3_8b_instruct_remove_40_seed3_jbbmixsft

swe_original-qwen-7b-25k-3epochs-5e-5

svd-safety-mistral_remove50_swapgapiter_rankunit_b010

Minecraft-Agent-Qwen3.5-9B-Stage3

svd-safety-gemma2_9b_it_swift_remove30

svd-safety-l2_swift_jbbsft1_remove50

SonoReasoner-8B

VERVE-Qwen3.5-9B-plain

rethink_rlvr_reproduce-ground_truth-qwen2.5_math_7b-lr5e-7-kl0.00-step50

rethink_rlvr_reproduce-ground_truth-qwen2.5_math_7b-lr5e-7-kl0.00-step100

Llama-3.1-8B-Instruct-DPO-100R0L-PoliTune

Hope-Models3

testEvan

llama_2_sky_safe_o1_llama_3_8B_reflect_4000_500_full

llama_2_llama_2_code_math_5_full

llama_2_cot_simplest_alpaca_4_full

camel

Llama-3.1-8B-Instruct_SFT_Math-220kv00.08

chat_bot_merged

Quelix-8B-v0.1

qwen2-5-7b-full-pretrain-mix-low-tweet-1m-en-reproduce-bs8

qwen2-5-7b-full-pretrain-mix-high-tweet-1m-en-reproduce-bs8

qwen3-8b-karma-v3-mlx-fp16

KD-Tinker

TIM-8b-preview