Mikhail's picture

Mikhail

RefalMachine

·

RefalMachine

AI & ML interests

NLP

Recent Activity

new activity 7 days ago

RefalMachine/RuadaptQwen3-32B-Instruct:Please add 1.7B model for speculative decoding

new activity 10 days ago

RefalMachine/RuadaptQwen3-32B-Instruct:Галлюцинации

updated a dataset 16 days ago

RefalMachine/darumeru

View all activity

Organizations

RefalMachine 's models 178

RefalMachine/RuadaptQwen2.5-Coder-1.5B-instruct_v2

Updated Nov 15, 2024 • 18

RefalMachine/ruadapt_qwen2.5_3B_ext_u32_instruct_v3

Updated Oct 17, 2024 • 8

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr5e4_peft_mlp_32_32_bs256

Updated Oct 16, 2024 • 3

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr3e4_peft_bs256_mlp

Updated Oct 16, 2024 • 64

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr5e4_peft_bs256

Updated Oct 16, 2024 • 6

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr3e4_peft_bs256_v2

Updated Oct 15, 2024 • 54

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr1e4_peft_bs256_v2

Updated Oct 15, 2024 • 42

RefalMachine/ruadapt_qwen_2.5_3B_ext_u32_lep_ft_sft_kto_v2

Updated Oct 15, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr1e4_peft_bs256

Updated Oct 15, 2024 • 41

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr5e4_bs256

Text Generation • Updated Oct 15, 2024 • 17

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_full_lr3e4_bs256

Text Generation • Updated Oct 14, 2024 • 15

RefalMachine/ruadapt_qwen2.5_3B_u48_full_lr3e4_bs256

Updated Oct 14, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_u48_mean_init

Updated Oct 14, 2024 • 10

RefalMachine/ruadapt_qwen2.5_3B_ext_u48_mean_init

Updated Oct 14, 2024 • 8 • 1

RefalMachine/ruadapt_qwen_2.5_3B_ext_u32_lep_ft_sft_kto

Updated Oct 13, 2024 • 8

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_unigram_32000_full_lr5e4_bs256

Updated Oct 13, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_unigram_32000_full_lr5e4_bs256

Updated Oct 12, 2024 • 5

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_bpe_32000_full_lr5e4_bs256

Updated Oct 12, 2024 • 4

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_bpe_32000_full_lr5e4_2k_bs256

Updated Oct 11, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_bpe_32000_full_lr3e4_2k_bs256

Updated Oct 11, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_bpe_32000_full_lr2e4_2k_bs256

Updated Oct 11, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_unigram_32000_full_lr3e4_bs256

Updated Oct 10, 2024 • 5

RefalMachine/ruadapt_qwen2.5_3B_bpe_32000_full_lr3e4_bs256

Updated Oct 10, 2024 • 7 • 1

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_unigram_32000_full_lr3e4_bs256

Updated Oct 10, 2024 • 6

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_bpe_32000_full_lr3e4_bs256

Updated Oct 10, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_unigram_32000_full_lr2e4_bs256

Updated Oct 10, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_unigram_32000_full_lr2e4_bs256

Updated Oct 10, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_ext_cl100k_bpe_32000_full_lr2e4_bs256

Updated Oct 10, 2024 • 6

RefalMachine/ruadapt_qwen2.5_3B_bpe_32000_full_lr2e4_bs256

Updated Oct 9, 2024 • 7

RefalMachine/ruadapt_qwen2.5_3B_darulm_bpe_64000_full_lr3e4_bs256

Updated Oct 9, 2024 • 7