Commit
•
081496b
1
Parent(s):
ef82163
Training in progress, step 4500, checkpoint
Browse files- checkpoint-4500/added_tokens.json +4 -0
- checkpoint-4500/config.json +92 -0
- checkpoint-4500/generation_config.json +9 -0
- checkpoint-4500/model.safetensors +3 -0
- checkpoint-4500/optimizer.pt +3 -0
- checkpoint-4500/preprocessor_config.json +19 -0
- checkpoint-4500/rng_state.pth +3 -0
- checkpoint-4500/scheduler.pt +3 -0
- checkpoint-4500/special_tokens_map.json +13 -0
- checkpoint-4500/spm_char.model +3 -0
- checkpoint-4500/tokenizer_config.json +63 -0
- checkpoint-4500/trainer_state.json +735 -0
- checkpoint-4500/training_args.bin +3 -0
checkpoint-4500/added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<ctc_blank>": 80,
|
3 |
+
"<mask>": 79
|
4 |
+
}
|
checkpoint-4500/config.json
ADDED
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/speecht5_tts",
|
3 |
+
"activation_dropout": 0.1,
|
4 |
+
"apply_spec_augment": true,
|
5 |
+
"architectures": [
|
6 |
+
"SpeechT5ForTextToSpeech"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"bos_token_id": 0,
|
10 |
+
"conv_bias": false,
|
11 |
+
"conv_dim": [
|
12 |
+
512,
|
13 |
+
512,
|
14 |
+
512,
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512
|
19 |
+
],
|
20 |
+
"conv_kernel": [
|
21 |
+
10,
|
22 |
+
3,
|
23 |
+
3,
|
24 |
+
3,
|
25 |
+
3,
|
26 |
+
2,
|
27 |
+
2
|
28 |
+
],
|
29 |
+
"conv_stride": [
|
30 |
+
5,
|
31 |
+
2,
|
32 |
+
2,
|
33 |
+
2,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2
|
37 |
+
],
|
38 |
+
"decoder_attention_heads": 12,
|
39 |
+
"decoder_ffn_dim": 3072,
|
40 |
+
"decoder_layerdrop": 0.1,
|
41 |
+
"decoder_layers": 6,
|
42 |
+
"decoder_start_token_id": 2,
|
43 |
+
"encoder_attention_heads": 12,
|
44 |
+
"encoder_ffn_dim": 3072,
|
45 |
+
"encoder_layerdrop": 0.1,
|
46 |
+
"encoder_layers": 12,
|
47 |
+
"encoder_max_relative_position": 160,
|
48 |
+
"eos_token_id": 2,
|
49 |
+
"feat_extract_activation": "gelu",
|
50 |
+
"feat_extract_norm": "group",
|
51 |
+
"feat_proj_dropout": 0.0,
|
52 |
+
"guided_attention_loss_num_heads": 2,
|
53 |
+
"guided_attention_loss_scale": 10.0,
|
54 |
+
"guided_attention_loss_sigma": 0.4,
|
55 |
+
"hidden_act": "gelu",
|
56 |
+
"hidden_dropout": 0.1,
|
57 |
+
"hidden_size": 768,
|
58 |
+
"initializer_range": 0.02,
|
59 |
+
"is_encoder_decoder": true,
|
60 |
+
"layer_norm_eps": 1e-05,
|
61 |
+
"mask_feature_length": 10,
|
62 |
+
"mask_feature_min_masks": 0,
|
63 |
+
"mask_feature_prob": 0.0,
|
64 |
+
"mask_time_length": 10,
|
65 |
+
"mask_time_min_masks": 2,
|
66 |
+
"mask_time_prob": 0.05,
|
67 |
+
"max_length": 1876,
|
68 |
+
"max_speech_positions": 1876,
|
69 |
+
"max_text_positions": 600,
|
70 |
+
"model_type": "speecht5",
|
71 |
+
"num_conv_pos_embedding_groups": 16,
|
72 |
+
"num_conv_pos_embeddings": 128,
|
73 |
+
"num_feat_extract_layers": 7,
|
74 |
+
"num_mel_bins": 80,
|
75 |
+
"pad_token_id": 1,
|
76 |
+
"positional_dropout": 0.1,
|
77 |
+
"reduction_factor": 2,
|
78 |
+
"scale_embedding": false,
|
79 |
+
"speaker_embedding_dim": 512,
|
80 |
+
"speech_decoder_postnet_dropout": 0.5,
|
81 |
+
"speech_decoder_postnet_kernel": 5,
|
82 |
+
"speech_decoder_postnet_layers": 5,
|
83 |
+
"speech_decoder_postnet_units": 256,
|
84 |
+
"speech_decoder_prenet_dropout": 0.5,
|
85 |
+
"speech_decoder_prenet_layers": 2,
|
86 |
+
"speech_decoder_prenet_units": 256,
|
87 |
+
"torch_dtype": "float32",
|
88 |
+
"transformers_version": "4.41.0.dev0",
|
89 |
+
"use_cache": false,
|
90 |
+
"use_guided_attention_loss": true,
|
91 |
+
"vocab_size": 81
|
92 |
+
}
|
checkpoint-4500/generation_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"decoder_start_token_id": 2,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"max_length": 1876,
|
7 |
+
"pad_token_id": 1,
|
8 |
+
"transformers_version": "4.41.0.dev0"
|
9 |
+
}
|
checkpoint-4500/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:98dd2696a28880760533cee43d3d20c83b7a845f585e69d01e697706ad39d6d4
|
3 |
+
size 577789320
|
checkpoint-4500/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6b33cada50a2d5be1158a532a42b4fd1c6f3e7dd6412ea2a9bfde73cb31bde67
|
3 |
+
size 1155772233
|
checkpoint-4500/preprocessor_config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": false,
|
3 |
+
"feature_extractor_type": "SpeechT5FeatureExtractor",
|
4 |
+
"feature_size": 1,
|
5 |
+
"fmax": 7600,
|
6 |
+
"fmin": 80,
|
7 |
+
"frame_signal_scale": 1.0,
|
8 |
+
"hop_length": 16,
|
9 |
+
"mel_floor": 1e-10,
|
10 |
+
"num_mel_bins": 80,
|
11 |
+
"padding_side": "right",
|
12 |
+
"padding_value": 0.0,
|
13 |
+
"processor_class": "SpeechT5Processor",
|
14 |
+
"reduction_factor": 2,
|
15 |
+
"return_attention_mask": true,
|
16 |
+
"sampling_rate": 16000,
|
17 |
+
"win_function": "hann_window",
|
18 |
+
"win_length": 64
|
19 |
+
}
|
checkpoint-4500/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b6a57f50f22137f48530075f96aa21d1395970102f476bb5173d2a65a3f6173
|
3 |
+
size 14244
|
checkpoint-4500/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1d9ea6ac8e84dcaff37314a282ec7aba7995babdb11261372593bc800ef835cf
|
3 |
+
size 1064
|
checkpoint-4500/special_tokens_map.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"eos_token": "</s>",
|
4 |
+
"mask_token": {
|
5 |
+
"content": "<mask>",
|
6 |
+
"lstrip": true,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false
|
10 |
+
},
|
11 |
+
"pad_token": "<pad>",
|
12 |
+
"unk_token": "<unk>"
|
13 |
+
}
|
checkpoint-4500/spm_char.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
|
3 |
+
size 238473
|
checkpoint-4500/tokenizer_config.json
ADDED
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"79": {
|
36 |
+
"content": "<mask>",
|
37 |
+
"lstrip": true,
|
38 |
+
"normalized": true,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"80": {
|
44 |
+
"content": "<ctc_blank>",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": true,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": false
|
50 |
+
}
|
51 |
+
},
|
52 |
+
"bos_token": "<s>",
|
53 |
+
"clean_up_tokenization_spaces": true,
|
54 |
+
"eos_token": "</s>",
|
55 |
+
"mask_token": "<mask>",
|
56 |
+
"model_max_length": 600,
|
57 |
+
"normalize": false,
|
58 |
+
"pad_token": "<pad>",
|
59 |
+
"processor_class": "SpeechT5Processor",
|
60 |
+
"sp_model_kwargs": {},
|
61 |
+
"tokenizer_class": "SpeechT5Tokenizer",
|
62 |
+
"unk_token": "<unk>"
|
63 |
+
}
|
checkpoint-4500/trainer_state.json
ADDED
@@ -0,0 +1,735 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.3881785571575165,
|
3 |
+
"best_model_checkpoint": "mikhail-panzo/zlm_b64_le5_s8000/checkpoint-4500",
|
4 |
+
"epoch": 3.769633507853403,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.041884816753926704,
|
13 |
+
"grad_norm": 11.218673706054688,
|
14 |
+
"learning_rate": 2.4500000000000004e-07,
|
15 |
+
"loss": 1.1196,
|
16 |
+
"step": 50
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.08376963350785341,
|
20 |
+
"grad_norm": 5.423076629638672,
|
21 |
+
"learning_rate": 4.95e-07,
|
22 |
+
"loss": 1.057,
|
23 |
+
"step": 100
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.1256544502617801,
|
27 |
+
"grad_norm": 12.713382720947266,
|
28 |
+
"learning_rate": 7.4e-07,
|
29 |
+
"loss": 1.0037,
|
30 |
+
"step": 150
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.16753926701570682,
|
34 |
+
"grad_norm": 4.434875965118408,
|
35 |
+
"learning_rate": 9.9e-07,
|
36 |
+
"loss": 0.9381,
|
37 |
+
"step": 200
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.2094240837696335,
|
41 |
+
"grad_norm": 3.668553590774536,
|
42 |
+
"learning_rate": 1.2400000000000002e-06,
|
43 |
+
"loss": 0.8591,
|
44 |
+
"step": 250
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.2513089005235602,
|
48 |
+
"grad_norm": 2.2456483840942383,
|
49 |
+
"learning_rate": 1.4900000000000001e-06,
|
50 |
+
"loss": 0.8262,
|
51 |
+
"step": 300
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.2931937172774869,
|
55 |
+
"grad_norm": 2.70170259475708,
|
56 |
+
"learning_rate": 1.74e-06,
|
57 |
+
"loss": 0.7634,
|
58 |
+
"step": 350
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.33507853403141363,
|
62 |
+
"grad_norm": 2.9946064949035645,
|
63 |
+
"learning_rate": 1.9900000000000004e-06,
|
64 |
+
"loss": 0.7336,
|
65 |
+
"step": 400
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.3769633507853403,
|
69 |
+
"grad_norm": 2.2950398921966553,
|
70 |
+
"learning_rate": 2.24e-06,
|
71 |
+
"loss": 0.7121,
|
72 |
+
"step": 450
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.418848167539267,
|
76 |
+
"grad_norm": 8.007049560546875,
|
77 |
+
"learning_rate": 2.4900000000000003e-06,
|
78 |
+
"loss": 0.6879,
|
79 |
+
"step": 500
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.418848167539267,
|
83 |
+
"eval_loss": 0.5903987884521484,
|
84 |
+
"eval_runtime": 258.509,
|
85 |
+
"eval_samples_per_second": 32.838,
|
86 |
+
"eval_steps_per_second": 4.108,
|
87 |
+
"step": 500
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.4607329842931937,
|
91 |
+
"grad_norm": 3.0767672061920166,
|
92 |
+
"learning_rate": 2.7400000000000004e-06,
|
93 |
+
"loss": 0.6734,
|
94 |
+
"step": 550
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.5026178010471204,
|
98 |
+
"grad_norm": 8.150467872619629,
|
99 |
+
"learning_rate": 2.99e-06,
|
100 |
+
"loss": 0.6539,
|
101 |
+
"step": 600
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.5445026178010471,
|
105 |
+
"grad_norm": 2.2473220825195312,
|
106 |
+
"learning_rate": 3.2400000000000003e-06,
|
107 |
+
"loss": 0.657,
|
108 |
+
"step": 650
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 0.5863874345549738,
|
112 |
+
"grad_norm": 2.448143482208252,
|
113 |
+
"learning_rate": 3.49e-06,
|
114 |
+
"loss": 0.6258,
|
115 |
+
"step": 700
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.6282722513089005,
|
119 |
+
"grad_norm": 4.810636520385742,
|
120 |
+
"learning_rate": 3.74e-06,
|
121 |
+
"loss": 0.6155,
|
122 |
+
"step": 750
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.6701570680628273,
|
126 |
+
"grad_norm": 2.5927581787109375,
|
127 |
+
"learning_rate": 3.990000000000001e-06,
|
128 |
+
"loss": 0.6073,
|
129 |
+
"step": 800
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.7120418848167539,
|
133 |
+
"grad_norm": 1.913765549659729,
|
134 |
+
"learning_rate": 4.24e-06,
|
135 |
+
"loss": 0.6118,
|
136 |
+
"step": 850
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.7539267015706806,
|
140 |
+
"grad_norm": 2.210376262664795,
|
141 |
+
"learning_rate": 4.49e-06,
|
142 |
+
"loss": 0.5807,
|
143 |
+
"step": 900
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 0.7958115183246073,
|
147 |
+
"grad_norm": 3.323096752166748,
|
148 |
+
"learning_rate": 4.74e-06,
|
149 |
+
"loss": 0.5915,
|
150 |
+
"step": 950
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.837696335078534,
|
154 |
+
"grad_norm": 2.1455421447753906,
|
155 |
+
"learning_rate": 4.9900000000000005e-06,
|
156 |
+
"loss": 0.5606,
|
157 |
+
"step": 1000
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 0.837696335078534,
|
161 |
+
"eval_loss": 0.49923568964004517,
|
162 |
+
"eval_runtime": 260.1008,
|
163 |
+
"eval_samples_per_second": 32.637,
|
164 |
+
"eval_steps_per_second": 4.083,
|
165 |
+
"step": 1000
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.8795811518324608,
|
169 |
+
"grad_norm": 2.7230141162872314,
|
170 |
+
"learning_rate": 5.240000000000001e-06,
|
171 |
+
"loss": 0.5731,
|
172 |
+
"step": 1050
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 0.9214659685863874,
|
176 |
+
"grad_norm": 2.0273685455322266,
|
177 |
+
"learning_rate": 5.490000000000001e-06,
|
178 |
+
"loss": 0.5361,
|
179 |
+
"step": 1100
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.9633507853403142,
|
183 |
+
"grad_norm": 3.9353837966918945,
|
184 |
+
"learning_rate": 5.74e-06,
|
185 |
+
"loss": 0.5476,
|
186 |
+
"step": 1150
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 1.0052356020942408,
|
190 |
+
"grad_norm": 2.7838985919952393,
|
191 |
+
"learning_rate": 5.99e-06,
|
192 |
+
"loss": 0.5356,
|
193 |
+
"step": 1200
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 1.0471204188481675,
|
197 |
+
"grad_norm": 2.8621716499328613,
|
198 |
+
"learning_rate": 6.24e-06,
|
199 |
+
"loss": 0.5413,
|
200 |
+
"step": 1250
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 1.0890052356020943,
|
204 |
+
"grad_norm": 2.490875005722046,
|
205 |
+
"learning_rate": 6.4900000000000005e-06,
|
206 |
+
"loss": 0.5375,
|
207 |
+
"step": 1300
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 1.130890052356021,
|
211 |
+
"grad_norm": 3.052311897277832,
|
212 |
+
"learning_rate": 6.740000000000001e-06,
|
213 |
+
"loss": 0.5415,
|
214 |
+
"step": 1350
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 1.1727748691099475,
|
218 |
+
"grad_norm": 2.5695457458496094,
|
219 |
+
"learning_rate": 6.99e-06,
|
220 |
+
"loss": 0.5299,
|
221 |
+
"step": 1400
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.2146596858638743,
|
225 |
+
"grad_norm": 2.5093400478363037,
|
226 |
+
"learning_rate": 7.24e-06,
|
227 |
+
"loss": 0.5175,
|
228 |
+
"step": 1450
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 1.256544502617801,
|
232 |
+
"grad_norm": 3.005453586578369,
|
233 |
+
"learning_rate": 7.49e-06,
|
234 |
+
"loss": 0.5203,
|
235 |
+
"step": 1500
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 1.256544502617801,
|
239 |
+
"eval_loss": 0.45439228415489197,
|
240 |
+
"eval_runtime": 266.4154,
|
241 |
+
"eval_samples_per_second": 31.864,
|
242 |
+
"eval_steps_per_second": 3.986,
|
243 |
+
"step": 1500
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.2984293193717278,
|
247 |
+
"grad_norm": 3.572266101837158,
|
248 |
+
"learning_rate": 7.74e-06,
|
249 |
+
"loss": 0.5192,
|
250 |
+
"step": 1550
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 1.3403141361256545,
|
254 |
+
"grad_norm": 2.545226573944092,
|
255 |
+
"learning_rate": 7.990000000000001e-06,
|
256 |
+
"loss": 0.5128,
|
257 |
+
"step": 1600
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.3821989528795813,
|
261 |
+
"grad_norm": 2.4371402263641357,
|
262 |
+
"learning_rate": 8.24e-06,
|
263 |
+
"loss": 0.5118,
|
264 |
+
"step": 1650
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 1.4240837696335078,
|
268 |
+
"grad_norm": 2.7607970237731934,
|
269 |
+
"learning_rate": 8.49e-06,
|
270 |
+
"loss": 0.5059,
|
271 |
+
"step": 1700
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 1.4659685863874345,
|
275 |
+
"grad_norm": 2.4732346534729004,
|
276 |
+
"learning_rate": 8.740000000000001e-06,
|
277 |
+
"loss": 0.5129,
|
278 |
+
"step": 1750
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 1.5078534031413613,
|
282 |
+
"grad_norm": 2.1999361515045166,
|
283 |
+
"learning_rate": 8.99e-06,
|
284 |
+
"loss": 0.4977,
|
285 |
+
"step": 1800
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 1.5497382198952878,
|
289 |
+
"grad_norm": 2.2437193393707275,
|
290 |
+
"learning_rate": 9.240000000000001e-06,
|
291 |
+
"loss": 0.4964,
|
292 |
+
"step": 1850
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 1.5916230366492146,
|
296 |
+
"grad_norm": 1.9161577224731445,
|
297 |
+
"learning_rate": 9.49e-06,
|
298 |
+
"loss": 0.4914,
|
299 |
+
"step": 1900
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.6335078534031413,
|
303 |
+
"grad_norm": 3.3452305793762207,
|
304 |
+
"learning_rate": 9.74e-06,
|
305 |
+
"loss": 0.4971,
|
306 |
+
"step": 1950
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 1.675392670157068,
|
310 |
+
"grad_norm": 1.818142294883728,
|
311 |
+
"learning_rate": 9.990000000000001e-06,
|
312 |
+
"loss": 0.4873,
|
313 |
+
"step": 2000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 1.675392670157068,
|
317 |
+
"eval_loss": 0.43226656317710876,
|
318 |
+
"eval_runtime": 264.8191,
|
319 |
+
"eval_samples_per_second": 32.056,
|
320 |
+
"eval_steps_per_second": 4.01,
|
321 |
+
"step": 2000
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 1.7172774869109948,
|
325 |
+
"grad_norm": 2.1289865970611572,
|
326 |
+
"learning_rate": 9.920000000000002e-06,
|
327 |
+
"loss": 0.4809,
|
328 |
+
"step": 2050
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 1.7591623036649215,
|
332 |
+
"grad_norm": 3.333080768585205,
|
333 |
+
"learning_rate": 9.836666666666668e-06,
|
334 |
+
"loss": 0.4853,
|
335 |
+
"step": 2100
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.8010471204188483,
|
339 |
+
"grad_norm": 3.106902599334717,
|
340 |
+
"learning_rate": 9.753333333333335e-06,
|
341 |
+
"loss": 0.4841,
|
342 |
+
"step": 2150
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 1.8429319371727748,
|
346 |
+
"grad_norm": 3.021559238433838,
|
347 |
+
"learning_rate": 9.67e-06,
|
348 |
+
"loss": 0.4795,
|
349 |
+
"step": 2200
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 1.8848167539267016,
|
353 |
+
"grad_norm": 2.128567695617676,
|
354 |
+
"learning_rate": 9.586666666666667e-06,
|
355 |
+
"loss": 0.4784,
|
356 |
+
"step": 2250
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.9267015706806283,
|
360 |
+
"grad_norm": 2.3653769493103027,
|
361 |
+
"learning_rate": 9.503333333333334e-06,
|
362 |
+
"loss": 0.4792,
|
363 |
+
"step": 2300
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 1.9685863874345548,
|
367 |
+
"grad_norm": 2.409151792526245,
|
368 |
+
"learning_rate": 9.42e-06,
|
369 |
+
"loss": 0.4727,
|
370 |
+
"step": 2350
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 2.0104712041884816,
|
374 |
+
"grad_norm": 2.9087882041931152,
|
375 |
+
"learning_rate": 9.336666666666666e-06,
|
376 |
+
"loss": 0.4645,
|
377 |
+
"step": 2400
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 2.0523560209424083,
|
381 |
+
"grad_norm": 2.5785608291625977,
|
382 |
+
"learning_rate": 9.253333333333333e-06,
|
383 |
+
"loss": 0.4696,
|
384 |
+
"step": 2450
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 2.094240837696335,
|
388 |
+
"grad_norm": 2.625706672668457,
|
389 |
+
"learning_rate": 9.17e-06,
|
390 |
+
"loss": 0.4688,
|
391 |
+
"step": 2500
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 2.094240837696335,
|
395 |
+
"eval_loss": 0.41558554768562317,
|
396 |
+
"eval_runtime": 267.455,
|
397 |
+
"eval_samples_per_second": 31.74,
|
398 |
+
"eval_steps_per_second": 3.971,
|
399 |
+
"step": 2500
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 2.136125654450262,
|
403 |
+
"grad_norm": 3.3157095909118652,
|
404 |
+
"learning_rate": 9.086666666666667e-06,
|
405 |
+
"loss": 0.4684,
|
406 |
+
"step": 2550
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 2.1780104712041886,
|
410 |
+
"grad_norm": 1.7106587886810303,
|
411 |
+
"learning_rate": 9.003333333333334e-06,
|
412 |
+
"loss": 0.4721,
|
413 |
+
"step": 2600
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 2.2198952879581153,
|
417 |
+
"grad_norm": 2.8600311279296875,
|
418 |
+
"learning_rate": 8.920000000000001e-06,
|
419 |
+
"loss": 0.463,
|
420 |
+
"step": 2650
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 2.261780104712042,
|
424 |
+
"grad_norm": 2.0305252075195312,
|
425 |
+
"learning_rate": 8.836666666666668e-06,
|
426 |
+
"loss": 0.4648,
|
427 |
+
"step": 2700
|
428 |
+
},
|
429 |
+
{
|
430 |
+
"epoch": 2.303664921465969,
|
431 |
+
"grad_norm": 3.124579906463623,
|
432 |
+
"learning_rate": 8.753333333333333e-06,
|
433 |
+
"loss": 0.4664,
|
434 |
+
"step": 2750
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 2.345549738219895,
|
438 |
+
"grad_norm": 2.565577268600464,
|
439 |
+
"learning_rate": 8.67e-06,
|
440 |
+
"loss": 0.456,
|
441 |
+
"step": 2800
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 2.387434554973822,
|
445 |
+
"grad_norm": 1.6280083656311035,
|
446 |
+
"learning_rate": 8.588333333333334e-06,
|
447 |
+
"loss": 0.4615,
|
448 |
+
"step": 2850
|
449 |
+
},
|
450 |
+
{
|
451 |
+
"epoch": 2.4293193717277486,
|
452 |
+
"grad_norm": 2.195096969604492,
|
453 |
+
"learning_rate": 8.505e-06,
|
454 |
+
"loss": 0.4712,
|
455 |
+
"step": 2900
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 2.4712041884816753,
|
459 |
+
"grad_norm": 2.9661295413970947,
|
460 |
+
"learning_rate": 8.421666666666668e-06,
|
461 |
+
"loss": 0.4607,
|
462 |
+
"step": 2950
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 2.513089005235602,
|
466 |
+
"grad_norm": 1.7868503332138062,
|
467 |
+
"learning_rate": 8.338333333333335e-06,
|
468 |
+
"loss": 0.4613,
|
469 |
+
"step": 3000
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 2.513089005235602,
|
473 |
+
"eval_loss": 0.405597448348999,
|
474 |
+
"eval_runtime": 267.2992,
|
475 |
+
"eval_samples_per_second": 31.758,
|
476 |
+
"eval_steps_per_second": 3.973,
|
477 |
+
"step": 3000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 2.554973821989529,
|
481 |
+
"grad_norm": 2.4901626110076904,
|
482 |
+
"learning_rate": 8.255000000000001e-06,
|
483 |
+
"loss": 0.4522,
|
484 |
+
"step": 3050
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 2.5968586387434556,
|
488 |
+
"grad_norm": 2.0079450607299805,
|
489 |
+
"learning_rate": 8.171666666666668e-06,
|
490 |
+
"loss": 0.4503,
|
491 |
+
"step": 3100
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 2.6387434554973823,
|
495 |
+
"grad_norm": 2.885077714920044,
|
496 |
+
"learning_rate": 8.088333333333334e-06,
|
497 |
+
"loss": 0.4487,
|
498 |
+
"step": 3150
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 2.680628272251309,
|
502 |
+
"grad_norm": 1.8914357423782349,
|
503 |
+
"learning_rate": 8.005e-06,
|
504 |
+
"loss": 0.4544,
|
505 |
+
"step": 3200
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 2.7225130890052354,
|
509 |
+
"grad_norm": 1.7431243658065796,
|
510 |
+
"learning_rate": 7.921666666666667e-06,
|
511 |
+
"loss": 0.4491,
|
512 |
+
"step": 3250
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 2.7643979057591626,
|
516 |
+
"grad_norm": 2.154836893081665,
|
517 |
+
"learning_rate": 7.838333333333334e-06,
|
518 |
+
"loss": 0.4488,
|
519 |
+
"step": 3300
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 2.806282722513089,
|
523 |
+
"grad_norm": 3.063981056213379,
|
524 |
+
"learning_rate": 7.755000000000001e-06,
|
525 |
+
"loss": 0.4501,
|
526 |
+
"step": 3350
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 2.8481675392670156,
|
530 |
+
"grad_norm": 1.9389702081680298,
|
531 |
+
"learning_rate": 7.671666666666668e-06,
|
532 |
+
"loss": 0.449,
|
533 |
+
"step": 3400
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 2.8900523560209423,
|
537 |
+
"grad_norm": 1.8112986087799072,
|
538 |
+
"learning_rate": 7.588333333333334e-06,
|
539 |
+
"loss": 0.4482,
|
540 |
+
"step": 3450
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 2.931937172774869,
|
544 |
+
"grad_norm": 2.060755729675293,
|
545 |
+
"learning_rate": 7.505e-06,
|
546 |
+
"loss": 0.4454,
|
547 |
+
"step": 3500
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 2.931937172774869,
|
551 |
+
"eval_loss": 0.39764249324798584,
|
552 |
+
"eval_runtime": 270.8158,
|
553 |
+
"eval_samples_per_second": 31.346,
|
554 |
+
"eval_steps_per_second": 3.921,
|
555 |
+
"step": 3500
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 2.973821989528796,
|
559 |
+
"grad_norm": 1.6399081945419312,
|
560 |
+
"learning_rate": 7.421666666666667e-06,
|
561 |
+
"loss": 0.455,
|
562 |
+
"step": 3550
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 3.0157068062827226,
|
566 |
+
"grad_norm": 1.8340182304382324,
|
567 |
+
"learning_rate": 7.338333333333334e-06,
|
568 |
+
"loss": 0.4442,
|
569 |
+
"step": 3600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 3.0575916230366493,
|
573 |
+
"grad_norm": 2.0830767154693604,
|
574 |
+
"learning_rate": 7.255000000000001e-06,
|
575 |
+
"loss": 0.443,
|
576 |
+
"step": 3650
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 3.099476439790576,
|
580 |
+
"grad_norm": 1.8929067850112915,
|
581 |
+
"learning_rate": 7.171666666666667e-06,
|
582 |
+
"loss": 0.4424,
|
583 |
+
"step": 3700
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 3.141361256544503,
|
587 |
+
"grad_norm": 1.7542285919189453,
|
588 |
+
"learning_rate": 7.088333333333334e-06,
|
589 |
+
"loss": 0.4348,
|
590 |
+
"step": 3750
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 3.183246073298429,
|
594 |
+
"grad_norm": 1.7470309734344482,
|
595 |
+
"learning_rate": 7.005000000000001e-06,
|
596 |
+
"loss": 0.4407,
|
597 |
+
"step": 3800
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 3.225130890052356,
|
601 |
+
"grad_norm": 1.6802048683166504,
|
602 |
+
"learning_rate": 6.921666666666668e-06,
|
603 |
+
"loss": 0.4372,
|
604 |
+
"step": 3850
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 3.2670157068062826,
|
608 |
+
"grad_norm": 1.4893133640289307,
|
609 |
+
"learning_rate": 6.838333333333334e-06,
|
610 |
+
"loss": 0.4348,
|
611 |
+
"step": 3900
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 3.3089005235602094,
|
615 |
+
"grad_norm": 3.0007338523864746,
|
616 |
+
"learning_rate": 6.7550000000000005e-06,
|
617 |
+
"loss": 0.4427,
|
618 |
+
"step": 3950
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 3.350785340314136,
|
622 |
+
"grad_norm": 1.806954264640808,
|
623 |
+
"learning_rate": 6.6716666666666674e-06,
|
624 |
+
"loss": 0.4299,
|
625 |
+
"step": 4000
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 3.350785340314136,
|
629 |
+
"eval_loss": 0.3921334147453308,
|
630 |
+
"eval_runtime": 264.7584,
|
631 |
+
"eval_samples_per_second": 32.063,
|
632 |
+
"eval_steps_per_second": 4.011,
|
633 |
+
"step": 4000
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 3.392670157068063,
|
637 |
+
"grad_norm": 2.3148508071899414,
|
638 |
+
"learning_rate": 6.588333333333334e-06,
|
639 |
+
"loss": 0.4321,
|
640 |
+
"step": 4050
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 3.4345549738219896,
|
644 |
+
"grad_norm": 1.546831488609314,
|
645 |
+
"learning_rate": 6.505e-06,
|
646 |
+
"loss": 0.4331,
|
647 |
+
"step": 4100
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 3.4764397905759163,
|
651 |
+
"grad_norm": 2.161104679107666,
|
652 |
+
"learning_rate": 6.421666666666667e-06,
|
653 |
+
"loss": 0.4388,
|
654 |
+
"step": 4150
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 3.518324607329843,
|
658 |
+
"grad_norm": 1.6207386255264282,
|
659 |
+
"learning_rate": 6.338333333333334e-06,
|
660 |
+
"loss": 0.4352,
|
661 |
+
"step": 4200
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 3.5602094240837694,
|
665 |
+
"grad_norm": 2.1468095779418945,
|
666 |
+
"learning_rate": 6.255e-06,
|
667 |
+
"loss": 0.4358,
|
668 |
+
"step": 4250
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 3.6020942408376966,
|
672 |
+
"grad_norm": 2.104029417037964,
|
673 |
+
"learning_rate": 6.171666666666667e-06,
|
674 |
+
"loss": 0.4314,
|
675 |
+
"step": 4300
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 3.643979057591623,
|
679 |
+
"grad_norm": 3.058011293411255,
|
680 |
+
"learning_rate": 6.088333333333334e-06,
|
681 |
+
"loss": 0.4298,
|
682 |
+
"step": 4350
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 3.6858638743455496,
|
686 |
+
"grad_norm": 1.4571338891983032,
|
687 |
+
"learning_rate": 6.005000000000001e-06,
|
688 |
+
"loss": 0.4364,
|
689 |
+
"step": 4400
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 3.7277486910994764,
|
693 |
+
"grad_norm": 1.5542718172073364,
|
694 |
+
"learning_rate": 5.921666666666667e-06,
|
695 |
+
"loss": 0.432,
|
696 |
+
"step": 4450
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 3.769633507853403,
|
700 |
+
"grad_norm": 2.5827696323394775,
|
701 |
+
"learning_rate": 5.838333333333334e-06,
|
702 |
+
"loss": 0.436,
|
703 |
+
"step": 4500
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 3.769633507853403,
|
707 |
+
"eval_loss": 0.3881785571575165,
|
708 |
+
"eval_runtime": 272.3698,
|
709 |
+
"eval_samples_per_second": 31.167,
|
710 |
+
"eval_steps_per_second": 3.899,
|
711 |
+
"step": 4500
|
712 |
+
}
|
713 |
+
],
|
714 |
+
"logging_steps": 50,
|
715 |
+
"max_steps": 8000,
|
716 |
+
"num_input_tokens_seen": 0,
|
717 |
+
"num_train_epochs": 7,
|
718 |
+
"save_steps": 500,
|
719 |
+
"stateful_callbacks": {
|
720 |
+
"TrainerControl": {
|
721 |
+
"args": {
|
722 |
+
"should_epoch_stop": false,
|
723 |
+
"should_evaluate": false,
|
724 |
+
"should_log": false,
|
725 |
+
"should_save": true,
|
726 |
+
"should_training_stop": false
|
727 |
+
},
|
728 |
+
"attributes": {}
|
729 |
+
}
|
730 |
+
},
|
731 |
+
"total_flos": 4.028518340583264e+16,
|
732 |
+
"train_batch_size": 16,
|
733 |
+
"trial_name": null,
|
734 |
+
"trial_params": null
|
735 |
+
}
|
checkpoint-4500/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6cbb0736de458ba42fdcabf5316a533e48ea0601626bceabd95443adeda6296a
|
3 |
+
size 5304
|