1B Parameter LLMs — Page 73

merge_accfmt_MRL4096_ROLLOUT4_LR5e-7_w0.9_linear

merge_cosfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_linear

merge_lenfmt_MRL4096_ROLLOUT4_LR5e-7_w0.7_linear

merge_accfmt_MRL4096_ROLLOUT4_LR1e-6_w0.5_ties

merge_lenfmt_MRL4096_ROLLOUT4_LR1e-6_w0.5_ties

merge_lenfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_ties_density0.2

SmolLM3-SFT

SmolLM3-SFT-Second-Round

StudyAiv17

math_merge_linear_1.5B

gemma3-fine-tuned

WitchLM-1.5B

gajosep

Zindi_RAC-Qwen2.5-1.5B-Instruct-Think-16-bit

gemma-3-1b-it-gsm8k-structured-reasoning-grpo-stage-1

ShweYon-Qwen2.5-Burmese-1.5B-v1.0

refund-assistant

gemma-3-1b-it-preference_dataset_mixture2_and_safe_pku-Preference

SN383

BLAST_PROCESSING-3.2-1B

095e2067

bt_v2

Llama-3.2-1B-Instruct-unsup-crf-full-weight-no-adapters

FluffyTail

unlearn_tofu_Llama-3.2-1B-Instruct_forget10_AltPO_lr1e-05_beta0.1_alpha2_epoch5

unlearn_tofu_Llama-3.2-1B-Instruct_forget10_RMU_lr5e-05_layer15_scoeff10_epoch5

unlearn_tofu_Llama-3.2-1B-Instruct_forget10_SimNPO_lr2e-05_b3.5_a1_d1_g0.125_ep10

qwen2.5-1.5b-grpo-sgd-linear

Qwen2.5-1.5B-random-weights

HDP-1B

M-Thinker-1.5B-Iter2

llama-3.2-1B-code-merged

finetuned-llama-3.2-1b-it-merged

gemma-3-1b-it-heretic

sft-qwen2.5-math-1.5b_Second

Gemma3-Quiet.Hours-1B

hh_qwen1.5_drpo_laplace_fixed_beta

gemma-3-1b-it-ghigliottina-grpo-merged-ckpt564

SFT_Z_model

Qwen2.5-1.5b-leetcode-math-linear

llama-sft-proj-layers

finetuned_llama3.1_1b_ollama_safe