4-5B Parameter LLMs — Page 101

mhm_ties__merge_experiments_math_think_11_ties_d0p3_l1p0

qwen3-ITSM-ticket

gemma4_ab_base_4bit_total

Qwen3-4B-Instruct-2507-FT-loan-collection-merged_v3_15_jun_26

anon-4B-RL

claust_v3_79

dqnScience

anon-4b-rl-thinking

Gemma-3-4B-IT-GA-SynthDolly-r16alpha32-E3-S73

Affine-S2-5GCaU8QYSjuVDZqueNtfwupwYvZBExctLCi7ZcCmaHdFkHUB

my-brain-v1

qwen35-4b-iconclass-grpo-v5hyb

qwen35-4b-iconclass-codesonly-poc2

Qwen3-4B-Tooluse-RLSD-TR

Qwen3-4B-Thinking-2507-hqq-mxfp4-faked-bf16

Uni-IAD-R2-Qwen3.5-GRPO-answer_first_2

repo_07_20_004

text2cypher_lora_v4_raw

gemma-2b-brain-v1

gmp-kd3e-1-s70pct-lr5e-5_20260811_115604

qwen3.5-4b-search-r1-sft

gemma-4-e2b-it

sparsegpt-qwen3-4b-s50pct

dama-aibrain

Affine-ded-ftr

Affine-H1-5GdomxEXGLwZS9ic4BwBHZdbfMNy8vNbWg3Bdze3JdFp6J5E

seed0_sample30000_mmmlu_google-gemma-3-4b-pt_multi_1.0-1.0_1.0

c17

qwen3-4b-dpo-qwen-cot-_2-3_05_DPO

deal-extractor-4b-v2

instruct_math_LS

rt-sam.backdoor_81_lr3e-5_rho0.01

rt-sam.backdoor_9_lr3e-5_rho0.01

qwen3-4B-refiner-sft-step-3201

k10-lr5e-7-ema0.01-eopd0.8-sciknoweval_material_sensitive20pct-pos_gap20pct

z0406_rt_broad_RT_backdoor_0_lr1e-6

z0406_rt_broad_RT_backdoor_1_lr3e-6

TTRL-sciknoweval_physics-TTRL-Len-8k-grpo-014723

z0406_rt_ordinary_RT_quirk_1_lr1e-5

z0406_rt_ordinary_RT_quirk_1_lr3e-5

z0406_rt_ordinary_RT_quirk_0_lr2e-5

Gemma-3-4B-IT-ES-SynthDolly-1A-E8