| # python3 -m espnet2.bin.s2t_train --use_preprocessor true --bpemodel data/token_list/bpe_unigram50000/bpe.model --token_type bpe --token_list data/token_list/bpe_unigram50000/tokens.txt --non_linguistic_symbols none --cleaner none --g2p none --valid_data_path_and_name_and_type dump/raw/cgn_valid/wav.scp,speech,kaldi_ark --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/speech_shape --resume true --fold_length 80000 --output_dir exp/s2t_lora_2 --config conf/lora_medium_2.yaml --frontend_conf fs=16k --normalize=global_mvn --normalize_conf stats_file=/ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/feats_stats.npz --train_data_path_and_name_and_type dump/raw/cgn_train/wav.scp,speech,kaldi_ark --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/speech_shape --fold_length 150 --train_data_path_and_name_and_type dump/raw/cgn_train/text.prev,text_prev,text --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/text_prev_shape.bpe --fold_length 150 --train_data_path_and_name_and_type dump/raw/cgn_train/text.ctc,text_ctc,text --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/text_ctc_shape.bpe --fold_length 150 --train_data_path_and_name_and_type dump/raw/cgn_train/text,text,text --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/text_shape.bpe --valid_data_path_and_name_and_type dump/raw/cgn_valid/text.prev,text_prev,text --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/text_prev_shape.bpe --valid_data_path_and_name_and_type dump/raw/cgn_valid/text.ctc,text_ctc,text --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/text_ctc_shape.bpe --valid_data_path_and_name_and_type dump/raw/cgn_valid/text,text,text --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/text_shape.bpe --model_conf extract_feats_in_collect_stats=false --ngpu 1 --multiprocessing_distributed True |
| # Started at Thu May 22 22:41:03 EDT 2025 |
| # |
| /ocean/projects/cis210027p/pwang10/miniconda3/envs/rota/bin/python3 /ocean/projects/cis210027p/pwang10/rota/espnet/espnet2/bin/s2t_train.py --use_preprocessor true --bpemodel data/token_list/bpe_unigram50000/bpe.model --token_type bpe --token_list data/token_list/bpe_unigram50000/tokens.txt --non_linguistic_symbols none --cleaner none --g2p none --valid_data_path_and_name_and_type dump/raw/cgn_valid/wav.scp,speech,kaldi_ark --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/speech_shape --resume true --fold_length 80000 --output_dir exp/s2t_lora_2 --config conf/lora_medium_2.yaml --frontend_conf fs=16k --normalize=global_mvn --normalize_conf stats_file=/ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/feats_stats.npz --train_data_path_and_name_and_type dump/raw/cgn_train/wav.scp,speech,kaldi_ark --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/speech_shape --fold_length 150 --train_data_path_and_name_and_type dump/raw/cgn_train/text.prev,text_prev,text --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/text_prev_shape.bpe --fold_length 150 --train_data_path_and_name_and_type dump/raw/cgn_train/text.ctc,text_ctc,text --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/text_ctc_shape.bpe --fold_length 150 --train_data_path_and_name_and_type dump/raw/cgn_train/text,text,text --train_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/text_shape.bpe --valid_data_path_and_name_and_type dump/raw/cgn_valid/text.prev,text_prev,text --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/text_prev_shape.bpe --valid_data_path_and_name_and_type dump/raw/cgn_valid/text.ctc,text_ctc,text --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/text_ctc_shape.bpe --valid_data_path_and_name_and_type dump/raw/cgn_valid/text,text,text --valid_shape_file /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/text_shape.bpe --model_conf extract_feats_in_collect_stats=false --ngpu 1 --multiprocessing_distributed True |
| [w002] 2025-05-22 22:44:32,588 (s2t:462) INFO: Vocabulary size: 50002 |
| [w002] 2025-05-22 22:44:37,175 (e_branchformer_encoder:445) INFO: Gradient checkpoint layers: [] |
| [w002] 2025-05-22 22:44:37,647 (transformer_decoder:103) INFO: Gradient checkpoint layers: [] |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.embed.conv.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.embed.conv.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.embed.conv.2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.embed.conv.2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.embed.out.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.embed.out.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,280 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.0.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,281 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.1.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.2.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,282 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.3.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,283 (abs_task:1370) INFO: Setting encoder.encoders.4.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.4.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.4.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.4.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.5.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,284 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.6.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.7.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,285 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.8.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,286 (abs_task:1370) INFO: Setting encoder.encoders.9.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.9.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.10.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,287 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.11.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,288 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.12.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.12.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.12.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.12.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.12.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.13.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,289 (abs_task:1370) INFO: Setting encoder.encoders.14.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.14.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.15.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,290 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.16.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.channel_proj1.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.channel_proj1.0.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.csgu.norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.csgu.norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.csgu.conv.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.csgu.conv.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.channel_proj2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.cgmlp.channel_proj2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward_macaron.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward_macaron.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,291 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward_macaron.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.feed_forward_macaron.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_ff.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_ff.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_ff_macaron.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_ff_macaron.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_mha.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_mha.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_mlp.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_mlp.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_final.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.norm_final.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.depthwise_conv_fusion.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.depthwise_conv_fusion.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.merge_proj.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.encoders.17.merge_proj.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.after_norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,292 (abs_task:1370) INFO: Setting encoder.after_norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.embed.0.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.after_norm.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.after_norm.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.output_layer.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.output_layer.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,293 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.0.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.1.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,294 (abs_task:1370) INFO: Setting decoder.decoders.2.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.2.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.2.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.2.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.2.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.2.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.2.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.3.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.4.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,295 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.5.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.6.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,296 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.7.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.8.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,297 (abs_task:1370) INFO: Setting decoder.decoders.9.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.9.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.9.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.10.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.11.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,298 (abs_task:1370) INFO: Setting decoder.decoders.12.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.12.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.13.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,299 (abs_task:1370) INFO: Setting decoder.decoders.14.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.14.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.15.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.16.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,300 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.self_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_q.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_q.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_k.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_k.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_v.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_v.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_out.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.src_attn.linear_out.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.feed_forward.w_1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.feed_forward.w_1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.feed_forward.w_2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.feed_forward.w_2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.norm1.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.norm1.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.norm2.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.norm2.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.norm3.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,301 (abs_task:1370) INFO: Setting decoder.decoders.17.norm3.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:40,304 (abs_task:1370) INFO: Setting ctc.ctc_lo.weight.requires_grad = False |
| [w002] 2025-05-22 22:44:40,304 (abs_task:1370) INFO: Setting ctc.ctc_lo.bias.requires_grad = False |
| [w002] 2025-05-22 22:44:54,279 (abs_task:1398) INFO: pytorch.version=2.1.2, cuda.available=True, cudnn.version=8902, cudnn.benchmark=False, cudnn.deterministic=True |
| [w002] 2025-05-22 22:44:54,314 (abs_task:1399) INFO: Model structure: |
| ESPnetS2TModel( |
| (frontend): DefaultFrontend( |
| (stft): Stft(n_fft=512, win_length=400, hop_length=160, center=True, normalized=False, onesided=True) |
| (frontend): Frontend() |
| (logmel): LogMel(sr=16000, n_fft=512, n_mels=80, fmin=0, fmax=8000.0, htk=False) |
| ) |
| (specaug): SpecAug( |
| (freq_mask): MaskAlongAxis(mask_width_range=[0, 27], num_mask=2, axis=freq) |
| (time_mask): MaskAlongAxisVariableMaxWidth(mask_width_ratio_range=[0.0, 0.05], num_mask=10, axis=time) |
| ) |
| (normalize): GlobalMVN(stats_file=/ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium/train/feats_stats.npz, norm_means=True, norm_vars=True) |
| (encoder): EBranchformerEncoder( |
| (embed): Conv2dSubsampling( |
| (conv): Sequential( |
| (0): Conv2d(1, 1024, kernel_size=(3, 3), stride=(2, 2)) |
| (1): ReLU() |
| (2): Conv2d(1024, 1024, kernel_size=(3, 3), stride=(2, 2)) |
| (3): ReLU() |
| ) |
| (out): Sequential( |
| (0): Linear(in_features=19456, out_features=1024, bias=True) |
| (1): PositionalEncoding( |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| ) |
| ) |
| (encoders): MultiSequential( |
| (0): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (1): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (2): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (3): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (4): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (5): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (6): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (7): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (8): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (9): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (10): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (11): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (12): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (13): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (14): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (15): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (16): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (17): EBranchformerEncoderLayer( |
| (attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (cgmlp): ConvolutionalGatingMLP( |
| (channel_proj1): Sequential( |
| (0): Linear(in_features=1024, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| ) |
| (csgu): ConvolutionalSpatialGatingUnit( |
| (norm): LayerNorm((2048,), eps=1e-12, elementwise_affine=True) |
| (conv): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (act): Identity() |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (channel_proj2): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (feed_forward_macaron): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): Swish() |
| ) |
| (norm_ff): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_ff_macaron): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mha): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_mlp): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm_final): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (depthwise_conv_fusion): Conv1d(2048, 2048, kernel_size=(31,), stride=(1,), padding=(15,), groups=2048) |
| (merge_proj): Linear(in_features=2048, out_features=1024, bias=True) |
| ) |
| ) |
| (after_norm): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| ) |
| (decoder): TransformerDecoder( |
| (embed): Sequential( |
| (0): Embedding(50002, 1024) |
| (1): PositionalEncoding( |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| ) |
| (after_norm): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (output_layer): Linear(in_features=1024, out_features=50002, bias=True) |
| (decoders): MultiSequential( |
| (0): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (1): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (2): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (3): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (4): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (5): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (6): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (7): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (8): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (9): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (10): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (11): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (12): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (13): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (14): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (15): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (16): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| (17): DecoderLayer( |
| (self_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (src_attn): MultiHeadedAttention( |
| (linear_q): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_k): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_v): Linear(in_features=1024, out_features=1024, bias=True) |
| (linear_out): Linear(in_features=1024, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (q_norm): Identity() |
| (k_norm): Identity() |
| ) |
| (feed_forward): PositionwiseFeedForward( |
| (w_1): Linear(in_features=1024, out_features=4096, bias=True) |
| (w_2): Linear(in_features=4096, out_features=1024, bias=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| (activation): ReLU() |
| ) |
| (norm1): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm2): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (norm3): LayerNorm((1024,), eps=1e-12, elementwise_affine=True) |
| (dropout): Dropout(p=0.1, inplace=False) |
| ) |
| ) |
| ) |
| (criterion_att): LabelSmoothingLoss( |
| (criterion): KLDivLoss() |
| ) |
| (ctc): CTC( |
| (ctc_lo): Linear(in_features=1024, out_features=50002, bias=True) |
| (ctc_loss): CTCLoss() |
| ) |
| ) |
|
|
| Model summary: |
| Class Name: ESPnetS2TModel |
| Total Number of model parameters: 1.04 B |
| Number of trainable parameters: 21.13 M (2.0%) |
| Size: 84.52 MB |
| Type: torch.float32 |
| [w002] 2025-05-22 22:44:54,314 (abs_task:1402) INFO: Optimizer: |
| AdamW ( |
| Parameter Group 0 |
| amsgrad: False |
| betas: [0.9, 0.98] |
| capturable: False |
| differentiable: False |
| eps: 1e-06 |
| foreach: None |
| fused: None |
| initial_lr: 0.0001 |
| lr: 2.5e-08 |
| maximize: False |
| weight_decay: 0.0 |
| ) |
| [w002] 2025-05-22 22:44:54,314 (abs_task:1403) INFO: Scheduler: PiecewiseLinearWarmupLR(warmup_steps_list=[0, 10000, 50000], warmup_lr_list=[0.0, 0.00025, 0.0001]) |
| [w002] 2025-05-22 22:44:54,359 (abs_task:1412) INFO: Saving the configuration in exp/s2t_lora_2/config.yaml |
| [w002] 2025-05-22 22:44:54,936 (abs_task:1477) INFO: Loading pretrained params from /ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/owsm_v3.1_ebf/valid.total_count.ave_5best.till45epoch.pth |
| [w002] 2025-05-22 22:45:28,000 (abs_task:1826) INFO: [train] Batch sampler: UnsortedBatchSampler(N-batch=22430, batch_size=8, key_file=/ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//train/speech_shape, |
| [w002] 2025-05-22 22:45:28,071 (abs_task:1827) INFO: [train] mini-batch sizes summary: N-batch=22430, mean=8.0, min=8, max=8 |
| [w002] 2025-05-22 22:45:28,381 (read_text:31) INFO: keys_to_load is not None, only loading 179440 keys from dump/raw/cgn_train/text.prev |
| [w002] 2025-05-22 22:45:28,525 (read_text:31) INFO: keys_to_load is not None, only loading 179440 keys from dump/raw/cgn_train/text.ctc |
| [w002] 2025-05-22 22:45:29,421 (read_text:31) INFO: keys_to_load is not None, only loading 179440 keys from dump/raw/cgn_train/text |
| [w002] 2025-05-22 22:45:29,622 (s2t:444) INFO: Optional Data Names: ('text_prev', 'text_ctc', 'text_spk2', 'text_spk3', 'text_spk4') |
| [w002] 2025-05-22 22:45:29,625 (abs_task:1851) INFO: [train] dataset: |
| ESPnetDataset( |
| speech: {"path": "dump/raw/cgn_train/wav.scp", "type": "kaldi_ark"} |
| text_prev: {"path": "dump/raw/cgn_train/text.prev", "type": "text"} |
| text_ctc: {"path": "dump/raw/cgn_train/text.ctc", "type": "text"} |
| text: {"path": "dump/raw/cgn_train/text", "type": "text"} |
| preprocess: <espnet2.train.preprocessor.S2TPreprocessor object at 0x1515cb819c00>) |
| [w002] 2025-05-22 22:45:29,687 (abs_task:1826) INFO: [valid] Batch sampler: UnsortedBatchSampler(N-batch=3220, batch_size=8, key_file=/ocean/projects/cis210027p/pwang10/rota/espnet/egs2/owsm_v3.1/s2t1/exp/s2t_stats_raw_bpe50000_medium//valid/speech_shape, |
| [w002] 2025-05-22 22:45:29,687 (abs_task:1827) INFO: [valid] mini-batch sizes summary: N-batch=3220, mean=8.0, min=8, max=9 |
| [w002] 2025-05-22 22:45:29,771 (read_text:31) INFO: keys_to_load is not None, only loading 25766 keys from dump/raw/cgn_valid/text.prev |
| [w002] 2025-05-22 22:45:29,834 (read_text:31) INFO: keys_to_load is not None, only loading 25766 keys from dump/raw/cgn_valid/text.ctc |
| [w002] 2025-05-22 22:45:29,921 (read_text:31) INFO: keys_to_load is not None, only loading 25766 keys from dump/raw/cgn_valid/text |
| [w002] 2025-05-22 22:45:29,999 (s2t:444) INFO: Optional Data Names: ('text_prev', 'text_ctc', 'text_spk2', 'text_spk3', 'text_spk4') |
| [w002] 2025-05-22 22:45:29,999 (abs_task:1851) INFO: [valid] dataset: |
| ESPnetDataset( |
| speech: {"path": "dump/raw/cgn_valid/wav.scp", "type": "kaldi_ark"} |
| text_prev: {"path": "dump/raw/cgn_valid/text.prev", "type": "text"} |
| text_ctc: {"path": "dump/raw/cgn_valid/text.ctc", "type": "text"} |
| text: {"path": "dump/raw/cgn_valid/text", "type": "text"} |
| preprocess: <espnet2.train.preprocessor.S2TPreprocessor object at 0x1515c4107cd0>) |
| [w002] 2025-05-22 22:46:04,848 (trainer:177) INFO: The training was resumed using exp/s2t_lora_2/checkpoint.pth |
| [w002] 2025-05-22 22:46:20,806 (trainer:336) INFO: 15/15epoch started |
| Failed to import Flash Attention, using ESPnet default: /ocean/projects/cis210027p/pwang10/miniconda3/envs/rota/lib/python3.10/site-packages/flash_attn_2_cuda.cpython-310-x86_64-linux-gnu.so: undefined symbol: _ZNK3c105Error4whatEv |
| [w002] 2025-05-22 22:51:34,557 (trainer:811) INFO: 15epoch:train:1-750batch: iter_time=0.001, forward_time=0.197, loss_ctc=49.742, loss_att=25.102, acc=0.806, loss=8.124, backward_time=0.184, grad_norm=32.872, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.027, optim0_lr0=9.750e-05, train_time=1.672 |
| [w002] 2025-05-22 22:56:26,351 (trainer:811) INFO: 15epoch:train:751-1500batch: iter_time=1.298e-04, forward_time=0.175, loss_ctc=52.287, loss_att=26.429, acc=0.800, loss=8.547, backward_time=0.180, grad_norm=34.537, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.022, optim0_lr0=9.733e-05, train_time=1.556 |
| [w002] 2025-05-22 23:01:17,953 (trainer:811) INFO: 15epoch:train:1501-2250batch: iter_time=1.333e-04, forward_time=0.175, loss_ctc=51.277, loss_att=25.950, acc=0.801, loss=8.387, backward_time=0.180, grad_norm=34.314, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.022, optim0_lr0=9.716e-05, train_time=1.555 |
| [w002] 2025-05-22 23:06:09,683 (trainer:811) INFO: 15epoch:train:2251-3000batch: iter_time=1.437e-04, forward_time=0.175, loss_ctc=49.961, loss_att=25.153, acc=0.808, loss=8.149, backward_time=0.180, grad_norm=35.128, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.023, optim0_lr0=9.698e-05, train_time=1.556 |
| [w002] 2025-05-22 23:11:01,259 (trainer:811) INFO: 15epoch:train:3001-3750batch: iter_time=1.277e-04, forward_time=0.175, loss_ctc=51.247, loss_att=25.930, acc=0.805, loss=8.381, backward_time=0.180, grad_norm=37.690, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.022, optim0_lr0=9.681e-05, train_time=1.555 |
| [w002] 2025-05-22 23:15:52,875 (trainer:811) INFO: 15epoch:train:3751-4500batch: iter_time=1.238e-04, forward_time=0.175, loss_ctc=49.378, loss_att=25.030, acc=0.804, loss=8.083, backward_time=0.180, grad_norm=33.097, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.022, optim0_lr0=9.664e-05, train_time=1.555 |
| [w002] 2025-05-22 23:20:43,810 (trainer:811) INFO: 15epoch:train:4501-5250batch: iter_time=1.233e-04, forward_time=0.174, loss_ctc=50.606, loss_att=25.830, acc=0.802, loss=8.316, backward_time=0.180, grad_norm=34.740, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.022, optim0_lr0=9.648e-05, train_time=1.552 |
| [w002] 2025-05-22 23:25:35,977 (trainer:811) INFO: 15epoch:train:5251-6000batch: iter_time=1.320e-04, forward_time=0.175, loss_ctc=50.471, loss_att=25.533, acc=0.802, loss=8.254, backward_time=0.181, grad_norm=33.755, clip=100.000, loss_scale=4.398e+12, optim_step_time=0.022, optim0_lr0=9.631e-05, train_time=1.558 |
| [w002] 2025-05-22 23:30:30,202 (trainer:811) INFO: 15epoch:train:6001-6750batch: iter_time=1.404e-04, forward_time=0.176, loss_ctc=51.734, loss_att=26.233, acc=0.802, loss=8.471, backward_time=0.182, grad_norm=33.953, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.614e-05, train_time=1.569 |
| [w002] 2025-05-22 23:35:24,147 (trainer:811) INFO: 15epoch:train:6751-7500batch: iter_time=1.328e-04, forward_time=0.176, loss_ctc=50.516, loss_att=25.603, acc=0.804, loss=8.269, backward_time=0.182, grad_norm=34.583, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.024, optim0_lr0=9.597e-05, train_time=1.567 |
| [w002] 2025-05-22 23:40:17,996 (trainer:811) INFO: 15epoch:train:7501-8250batch: iter_time=1.320e-04, forward_time=0.176, loss_ctc=50.687, loss_att=25.559, acc=0.805, loss=8.274, backward_time=0.182, grad_norm=35.027, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.581e-05, train_time=1.567 |
| [w002] 2025-05-22 23:45:12,184 (trainer:811) INFO: 15epoch:train:8251-9000batch: iter_time=1.375e-04, forward_time=0.176, loss_ctc=50.193, loss_att=25.279, acc=0.801, loss=8.188, backward_time=0.182, grad_norm=34.127, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.564e-05, train_time=1.569 |
| [w002] 2025-05-22 23:50:06,919 (trainer:811) INFO: 15epoch:train:9001-9750batch: iter_time=1.334e-04, forward_time=0.176, loss_ctc=50.496, loss_att=25.463, acc=0.806, loss=8.243, backward_time=0.182, grad_norm=34.973, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.548e-05, train_time=1.572 |
| [w002] 2025-05-22 23:55:03,726 (trainer:811) INFO: 15epoch:train:9751-10500batch: iter_time=1.297e-04, forward_time=0.177, loss_ctc=50.950, loss_att=25.938, acc=0.802, loss=8.360, backward_time=0.183, grad_norm=35.053, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.025, optim0_lr0=9.532e-05, train_time=1.582 |
| [w002] 2025-05-22 23:59:58,438 (trainer:811) INFO: 15epoch:train:10501-11250batch: iter_time=1.331e-04, forward_time=0.176, loss_ctc=50.775, loss_att=25.845, acc=0.803, loss=8.331, backward_time=0.183, grad_norm=33.787, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.516e-05, train_time=1.572 |
| [w002] 2025-05-23 00:04:53,518 (trainer:811) INFO: 15epoch:train:11251-12000batch: iter_time=1.323e-04, forward_time=0.177, loss_ctc=52.155, loss_att=26.392, acc=0.800, loss=8.530, backward_time=0.183, grad_norm=34.980, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.500e-05, train_time=1.573 |
| [w002] 2025-05-23 00:09:48,668 (trainer:811) INFO: 15epoch:train:12001-12750batch: iter_time=1.293e-04, forward_time=0.177, loss_ctc=51.930, loss_att=26.329, acc=0.799, loss=8.502, backward_time=0.183, grad_norm=35.266, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.024, optim0_lr0=9.483e-05, train_time=1.574 |
| [w002] 2025-05-23 00:14:43,800 (trainer:811) INFO: 15epoch:train:12751-13500batch: iter_time=1.310e-04, forward_time=0.176, loss_ctc=50.843, loss_att=25.723, acc=0.802, loss=8.315, backward_time=0.183, grad_norm=34.735, clip=100.000, loss_scale=8.796e+12, optim_step_time=0.023, optim0_lr0=9.468e-05, train_time=1.574 |
| [w002] 2025-05-23 00:19:40,494 (trainer:811) INFO: 15epoch:train:13501-14250batch: iter_time=1.315e-04, forward_time=0.179, loss_ctc=49.959, loss_att=25.243, acc=0.807, loss=8.164, backward_time=0.183, grad_norm=33.037, clip=100.000, loss_scale=1.171e+13, optim_step_time=0.023, optim0_lr0=9.452e-05, train_time=1.583 |
| [w002] 2025-05-23 00:24:35,886 (trainer:811) INFO: 15epoch:train:14251-15000batch: iter_time=1.310e-04, forward_time=0.177, loss_ctc=50.325, loss_att=25.319, acc=0.805, loss=8.205, backward_time=0.183, grad_norm=34.725, clip=100.000, loss_scale=1.759e+13, optim_step_time=0.023, optim0_lr0=9.436e-05, train_time=1.575 |
| [w002] 2025-05-23 00:32:21,975 (trainer:394) INFO: 15epoch results: [train] iter_time=1.997e-04, forward_time=0.177, loss_ctc=50.776, loss_att=25.694, acc=0.803, loss=8.305, backward_time=0.182, grad_norm=34.519, clip=100.000, loss_scale=7.623e+12, optim_step_time=0.023, optim0_lr0=9.591e-05, train_time=1.572, time=1 hour, 38 minutes and 15.32 seconds, total_count=225000, gpu_max_cached_mem_GB=43.508, [valid] loss_ctc=18.456, cer_ctc=0.097, loss_att=9.018, acc=0.931, cer=0.049, wer=0.729, loss=11.849, time=7 minutes and 45.68 seconds, total_count=48300, gpu_max_cached_mem_GB=43.508 |
| [w002] 2025-05-23 00:32:47,576 (trainer:462) INFO: The best model has been updated: valid.acc, valid.total_count |
| [w002] 2025-05-23 00:32:47,578 (trainer:516) INFO: The model files were removed: exp/s2t_lora_2/14epoch.pth |
| [w002] 2025-05-23 00:32:47,578 (trainer:534) INFO: The training was finished at 15 epochs |
| # Accounting: time=6726 threads=1 |
| # Ended (code 0) at Fri May 23 00:33:09 EDT 2025, elapsed time 6726 seconds |
|
|
