2-3B Parameter LLMs — Page 22

qwen3-1.7b-coffee-sft

qwen25_3b_instruct_qwen25_qwen3_rank_only-qwen25_qwen3_rank_only_cluster_3

qwen25-3b-peacetalk-magic-v2-merged

Qwen3-1.7B-Base_dsum_3_6_1p0_0p0_1p0_grpo_sapo_42_rule

Qwen3-1.7B-Base_dsum_3_6_rel_1e-1_1p0_0p0_1p0_grpo_dr_grpo_42_rule

Qwen2.5-3B-Deconstruct-V2.4-Merged-v2

toolcalling-merged-demo

influence_metamath_qwen2.5-3b_repeat_regularized_1k_scaled_e1

Qwen2.5-3B-Instruct_adaptive_tune_no_ref

Qwen3-1.7B-Catgirl-test0430

toolcalling-merged-demo

17718_sft_16

Gemma-Daktari-Pawa_v2

pawa_math_grpo

gemma-2-2b-it_RMU_s400_a500_layer11

gemma-2-2b-it_RMU_s200_a500_layer15

6851_mcq_16_16

gemma-2-2b-it_RMU_s400_a1200_layer3

6851_mcq_64_16_new_format_combined

6851_mcq_32_32

Gradience-T1-3B-preview

Qwen3-1.7B-ShiningValiant3

EduRABSA_SLM_v1_SLERP_phi4mini

Qwen3-1.7B_modified

magistrate-3.2-3b-it

calme-3.1-baguette-3b

qwen-3-1.7b-finetuned

gemma-2b-it-edcastr_JavaScript-v3

qwen-2.5-3b-r1-countdown

Qwen3-1.7B-GRPO-SRT-Math-12k-Stage-1

Qwen2.5-3B-Instruct_unsloth_w_new_merged

Llama-3.2-3B-Instruct_unsloth_w_new_merged

qwen2.5-3b-dpo-coarse

orca_mini_v9_5_3B-Instruct

qwen3-1.7b-dspo-sft-base

OctoThinker-3B-Short-Base

reasoning-llama3.2-3b

LogicBench-Qwen-FT-Response

qwen3_1.7b_vanilla_psyscam_vanilla_romance

kenya-land-rights-gemma-2b-merged

gemma-2-2b-CoT-sft-thing-format-moredataset-sft2-fix

EvoNet-3B-V1