2-3B Parameter LLMs — Page 46

qwen3_1.7b_sudoku_multi_action_easy_21_30

north_llama32_3b_enhancedNCC_base_v1_lr1e5_2048_80000

qwen3_1.7b_sudoku_multi_action_easy_21_30_epoch1

qwen3_1.7b_new_sudoku_one_action_A_sft_lr_5e_6__step_562

qwen3_1.7b_new_sudoku_one_action_B_sft_lr_5e_6__step_4432

qwen3_1.7b_new_sudoku_one_action_C_sft_lr_5e_6__step_1668

qwen3_1.7b_sudoku_multi_action_easy_11_20_epoch2

qwen3_1.7b_sudoku_multi_action_easy_11_20_epoch1

qwen3_1.7b_sudoku_one_action_easy_11_20_epoch1

64b_RL_DAPO_step250

4b_SFT_NEW

self-debate-baseline-Qwen3-1.7B-Base-DAPO-n8-bs256-long8-step200

agentic-sudoku-NonMarkov_qwen2.5-3B-5e-6_9x9_6-6_gt-SFT_ans1-7k

CORE-Qwen3-1.7B-MATH-A9-U-S

qwen3-1.7b-base-adam-1e-6-bs128-kl0.0-global_step_20

qwen3-1.7b-base-svd-muon-adam-3e-6-bs128-kl0.0-global_step_200

qwen3-1.7b-base-svd-muon-adam-1e-6-bs128-kl0.0-global_step_100

qwen3-1.7b-base-svd-muon-adam-1e-6-bs128-kl0.0-global_step_120

qwen3-1.7b-base-svd-muon-adam-1e-6-bs128-kl0.0-global_step_140

qwen3-1.7b-base-svd-muon-adam-1e-6-bs128-kl0.0-global_step_160

qwen3_1.7b_rush_hour_one_move_4_9_epoch1

qwen3_1.7b_rush_hour_one_move_4_9_epoch2

qwen3_1.7b_rush_hour_one_move_4_9_epoch3

qwen3_1.7b_rush_hour_one_move_4_9

qwen3_1.7b_rush_hour_multi_move_final_10_12

qwen3_1.7b_sudoku_one_action_easy_21_30_epoch3

qwen3_1.7b_rush_hour_multi_move_final_short_4_9_epoch2

qwen25-3b-l3l3-ep5

nvidia_math_cot_1e5_v2_ep5

qwen3-1.7b-base-instruction-tuning-full-sft

tars-3b-merged

epstein-llama-3.2-3B

qwen3-1.7b-amr-augmented-20260214-1807

guinius-giro-003

qwen-25-3b-it-sft4500-len8192-rl-bs32-gs20

first_qwen3_1.7b

Chess-1.7B-v2

big-math-hard-tiny-qwen2.5-3b-instruct-og-rloo-implicit-cheat-no-global_step_45

test-e2e-qwen3-1.7b-fft-modal-test

qwen3_1.7b_sudoku_multi_action_group_norm_epoch2

dpo-llama-3.2-3b-set1-pref100

NL2SQL_finetuned