1B Parameter LLMs — Page 95

Sibatikgal

Finetuned

llama3-1b-gt-g-s-e

llama32_1bi_stdsft_rs0_3_5cut_e2

Priapus-3.2-1B

llama32_1bi_CoTsft_rs0_3_5cut_gem3_e2

1_layer_MQA_llama_model

Llama-3.2-1B-Instruct_sum_KTO_10k_1_1ep_4bit

10_layer_MQA_llama_model

tfa_output_2025_m05_d10_t20h_16m_03s

Llama-3.2-1B-Instruct_finetuned_1_lora

llamainstructgoodendings

test_model

Qwen2.5-Coder-1.5B-Instruct-Gensyn-Swarm-mammalian_smooth_bat

merge_linear_len0.3fmt0.7_MRL4096_ROLLOUT4_LR1e-6

merge_accfmt_MRL4096_ROLLOUT4_LR5e-7_w0.1_linear

merge_cosfmt_MRL4096_ROLLOUT4_LR2e-6_w0.9_linear

merge_lenfmt_MRL4096_ROLLOUT4_LR2e-6_w0.1_linear

merge_accfmt_MRL4096_ROLLOUT4_LR2e-6_w0.5_ties

merge_cosfmt_MRL4096_ROLLOUT4_LR2e-6_w0.5_dare_ties

merge_lenfmt_MRL4096_ROLLOUT4_LR2e-6_w0.5_ties

merge_accfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_dare_ties

merge_accfmt_MRL4096_ROLLOUT4_LR1e-6_w0.5_dare_ties

merge_cosfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_ties_density0.2

merge_lenfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_ties_density0.2

merge_lenfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_dare_ties_density0.2

qwen2.5-finetuned

Qwen2.5-Coder-1.5B-Instruct-Gensyn-Swarm-peaceful_sleek_bear

Qwen2.5-1.5B-Instruct-Gensyn-Swarm-reclusive_energetic_frog

StudyAiv22

Qwen2.5-1.5B-SFT-Schwinn

Ordis-1.5B-V355-VarGH

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

tinyllama-1.1B-sparse-10

Llama-3.2-1B-random-weights

GenRM-CI-Test-1.5B

temp-qwen2.5-1.5b-koeantextbook-finetuned

llama-converted-back

llama-sft-muon

chess-qwen2.5

chess-qwen-lora-v2

llama-sft-baseline