2-3B Parameter LLMs — Page 123

orpo-phi2

Qwen2.5-3B-Base-Math-v2

Qwen3-1.7B-teacher-refusal-integer

tft-benchmark-s1-tft-Qwen3-1.7B

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint500

Qwen3-1.7B

sft__ot30k_Qwen3-1.7B-Base-SFT-Tulu3-decontaminated

gemma-2b-it-noised-np0.25

gemma-2b-it-noised-np0.15-emb

SST-MetaxPyTorch-Hackathon-Merged16bit

Qwen-3B-Instruct-Vix-Exic

Qwen3-1.7B-Base-ftjob-a4c31a74a61b

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint275

Qwen3-1.7B-ultrachat-bsz128-ts300-regular-skywork8b-seed42-lr1e-6-warmup10-checkpoint100

Assignment3_Question1_qwen3-1.7b-backward-merged

gemma-2b-it-steer-dragon-numbers-ft

Qwen3-1.7B-icl-3shot-v4_128k

medmcqa-Qwen2.5-3B-finetuned

Qwen3-1.7B-Base-dapo_filter-grpo-useKL_True-KLlossCoef1e-3

KG-R1-CWQ-no-retrieval-reward

survey-xml-base-knowledge-0.0.1-merged_16bit

phi2-docstring-model1

qw3vl2b_evq_grp

Llama-3.2-3B-Instruct_grpo_ppl_adv_rollout_8_20260501_120104_step580

safety_model

PrAg-PO-Qwen3-1.7b-step720

math_skywork-v2-qwen3-1p7b-not_easy_1e-4_200

Qwen3-1.7B-nq-text-100k-with_pseudo_queries

group_model

general_knowledge_model

math_model

safety_model

multilingual_model

group_model

general_knowledge_model

group_model

qwen3-1.7b-summarization-arxiv-full

general_knowledge_model

qwen3-1.7b-summarization-mediasum

multilingual_model

math_model

math_model