4-5B Parameter LLMs — Page 107
4,593THChou1220ColdTools5B32K
gemma-4-e2b-kinetics54K-enhanced-fall_FFT
sagnikMColdTools4B32K
llama_qwen_nokl_posterior_step_100
yunjae-wonColdTools4B32K
T14b_8b_S4b_klreg5_checkpoint20
sagnikMColdTools4B32K
llama_qwen_prior_step_100
sagnikMColdTools4B32K
llama_qwen_nokl_prior_step_100
Stage-orgColdTools5B32K
appworld-qwen35-4b-qwen38-raw_source-epoch3-iter1
Stage-orgColdTools5B32K
appworld-qwen35-4b-luna-epoch1-iter1
Stage-orgColdTools5B32K
appworld-qwen35-4b-9b_info_gain_1-epoch2-iter1
surrey-nlpCold4B32KVision
oof-baroomfColdTools4B32K
csrsef-instruct-20260323T195339Z-it01-pubmedqa
HahmdongColdTools4B32K
AT-qwen3-4b-ultrachat-hhrlhf-15360-rm-ppo-clean-p0_05-step-50
blacksimon818ColdTools4B32K
bhavyagoyal-lexsiColdTools4B32K
harper-valley-qwen-sft-merged
hamishiviColdTools4B32K
fixed_rl_v3_tmax_combined_agent
robustness-smi-testsColdTools4B32K
rt-sam.backdoor_81_lr1e-5_rho0.01
robustness-smi-testsColdTools4B32K
rt-broad_RT.quirk_107_lr3e-5
hariharanv04ColdTools4B32K
qwen3-4b-instruct-3k-simple2
Johnny1024ColdTools4B32K
k10-lr5e-7-ema0.01-eopd0.8-sciknoweval_physics_sensitive20pct-pos_gap20pct
smi-robustness-eightColdTools4B32K
z0406_rt_broad_RT_backdoor_0_lr3e-5
smi-robustness-eightColdTools4B32K
z0406_rt_ordinary_RT_backdoor_0_lr3e-6
smi-robustness-eightColdTools4B32K
z0406_rt_broad_RT_quirk_0_lr1e-6
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_quirk_0_lr5e-5
smi-robustness-eightColdTools4B32K
z0406_rt_sam_RT_backdoor_0_lr3e-5_rho0.005
smi-robustness-eightColdTools4B32K
z0406_rt_sam_RT_backdoor_1_lr3e-5_rho0.005
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_quirk_1_lr2e-5
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_backdoor_0_lr5e-5
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_backdoor_0_lr2e-5
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_quirk_1_lr5e-5
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_quirk_1_lr1e-4
smi-robustness-bbibbiColdTools4B32K
z0406_rt_ordinary_RT_backdoor_0_lr1e-4
dinobbyColdTools4B32K
Qwen3-4B-Instruct-2507-Cog
Johnny1024ColdTools4B32K
bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-essay_bottom20_nogap-maxsteps150
Johnny1024ColdTools4B32K
bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_physics_bottom20_nogap-maxsteps150
Johnny1024ColdTools4B32K
intuitor-mmlu_pro-qwen3-4b-think-2507-r6k100
Johnny1024ColdTools4B32K
TTRL-essay-TTRL-Len-8k-grpo-024343
ArtemisnanaColdTools4B32K