2-3B Parameter LLMs — Page 53

qwen3_1.7b_easy_rl_final_step120

qwen3_1.7b_easy_rl_gspo

slm-hcmut

agentic-sokoban-Markov_qwen2.5-3B-it-5e-6_gt-SFT_6k

htktai2025-merged-model-v6

meta-llama_Llama-3.2-3B-Instruct-GRPO-vanilla_G_4-checkpoint-393

meta-llama_Llama-3.2-3B-Instruct-GRPO-vanilla_G_4-checkpoint-186

Llama-3.2-3B-Instruct_old_sft

PRM-llama3.2-3b-alpacafarm-sft

32b_RL

Vanilla_RL

qwen3_1.7b_easy_rl_ours_adv_fixed_gamma_1_98_geo_ms_token_tis

qwen-3B-stego-2-codes

2b_SFT_NEW

qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_96

Vanilla_RL_NEW

qwen3_1.7b_easy_rl_ours_adv_fixed_no_norm

qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_198

qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_396

qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_792

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_1184

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_296

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_888

qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_592

qwen3_1.7b_rush_hour_one_move_sft

qwen3_1.7b_sudoku_one_act_new

qwen3_1.7b_rush_hour_one_move_sft_new

qwen317step300

agentic-sudoku-qwen2.5-3B-it_SAS_SFT_pos

self-debate-exp-Qwen2.5-3B-grpo-diff_sol2048-n8-bs256-long8-DAPO-step200

4b_RL_DAPO

2b_RL_DAPO

1b_RL_DAPO

qwen3_1.7b_rush_hour_multi_move_final

agentic-sokoban-qwen2.5-3B_SAS_SFT

2b_SFT

1b_SFT

STaR_SFT

qwen3_1.7b_sudoku_one_action_easy_11_20_epoch2

qwen3_1.7b_sudoku_one_action_easy_11_20_epoch3

agentic-sudoku-Markov_qwen2.5-3B-5e-6_gt-SFT_ans1-24k

agentic-sudoku-NonMarkov_qwen2.5-3B-5e-6_gt-SFT_ans1-24k