Commit
•
2f8af2c
1
Parent(s):
4a22531
Training in progress, step 7500, checkpoint
Browse files- checkpoint-7500/added_tokens.json +4 -0
- checkpoint-7500/config.json +92 -0
- checkpoint-7500/generation_config.json +9 -0
- checkpoint-7500/model.safetensors +3 -0
- checkpoint-7500/optimizer.pt +3 -0
- checkpoint-7500/preprocessor_config.json +19 -0
- checkpoint-7500/rng_state.pth +3 -0
- checkpoint-7500/scheduler.pt +3 -0
- checkpoint-7500/special_tokens_map.json +13 -0
- checkpoint-7500/spm_char.model +3 -0
- checkpoint-7500/tokenizer_config.json +63 -0
- checkpoint-7500/trainer_state.json +1203 -0
- checkpoint-7500/training_args.bin +3 -0
checkpoint-7500/added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<ctc_blank>": 80,
|
3 |
+
"<mask>": 79
|
4 |
+
}
|
checkpoint-7500/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-7500/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-7500/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:112f40056d0e577bfc35c5857db5c7cb70e73509ef1228d590fae8bee497803f
|
3 |
+
size 577789320
|
checkpoint-7500/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d84bfbb3df655e81bcd83845b36f5ed3530e9adb21475193b74278fbcabecb4
|
3 |
+
size 1155772233
|
checkpoint-7500/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-7500/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:db0594178aff73d8a1c8ef8c6b307b40e29297118f04d166bb86bc745c050535
|
3 |
+
size 14244
|
checkpoint-7500/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ba9826396d1061aee6fdf7293fd789e47f8a5ac9db78d8396f43ecdcae9b12a2
|
3 |
+
size 1064
|
checkpoint-7500/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-7500/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-7500/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-7500/trainer_state.json
ADDED
@@ -0,0 +1,1203 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.3775756061077118,
|
3 |
+
"best_model_checkpoint": "mikhail-panzo/zlm_b64_le5_s8000/checkpoint-7500",
|
4 |
+
"epoch": 6.28140703517588,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 7500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04187604690117253,
|
13 |
+
"grad_norm": 5.415271759033203,
|
14 |
+
"learning_rate": 2.4000000000000003e-07,
|
15 |
+
"loss": 1.1819,
|
16 |
+
"step": 50
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.08375209380234507,
|
20 |
+
"grad_norm": 16.256921768188477,
|
21 |
+
"learning_rate": 4.900000000000001e-07,
|
22 |
+
"loss": 1.0035,
|
23 |
+
"step": 100
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.12562814070351758,
|
27 |
+
"grad_norm": 3.2439260482788086,
|
28 |
+
"learning_rate": 7.350000000000001e-07,
|
29 |
+
"loss": 0.9742,
|
30 |
+
"step": 150
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.16750418760469013,
|
34 |
+
"grad_norm": 4.24526834487915,
|
35 |
+
"learning_rate": 9.85e-07,
|
36 |
+
"loss": 0.9361,
|
37 |
+
"step": 200
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.20938023450586266,
|
41 |
+
"grad_norm": 3.625537872314453,
|
42 |
+
"learning_rate": 1.235e-06,
|
43 |
+
"loss": 0.8733,
|
44 |
+
"step": 250
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.25125628140703515,
|
48 |
+
"grad_norm": 4.1455464363098145,
|
49 |
+
"learning_rate": 1.485e-06,
|
50 |
+
"loss": 0.8471,
|
51 |
+
"step": 300
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.2931323283082077,
|
55 |
+
"grad_norm": 3.164158582687378,
|
56 |
+
"learning_rate": 1.7350000000000001e-06,
|
57 |
+
"loss": 0.838,
|
58 |
+
"step": 350
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.33500837520938026,
|
62 |
+
"grad_norm": 4.24931001663208,
|
63 |
+
"learning_rate": 1.985e-06,
|
64 |
+
"loss": 0.7467,
|
65 |
+
"step": 400
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.3768844221105528,
|
69 |
+
"grad_norm": 2.332688331604004,
|
70 |
+
"learning_rate": 2.235e-06,
|
71 |
+
"loss": 0.7389,
|
72 |
+
"step": 450
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.4187604690117253,
|
76 |
+
"grad_norm": 2.370961904525757,
|
77 |
+
"learning_rate": 2.4850000000000003e-06,
|
78 |
+
"loss": 0.7074,
|
79 |
+
"step": 500
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.4187604690117253,
|
83 |
+
"eval_loss": 0.6029486060142517,
|
84 |
+
"eval_runtime": 220.2253,
|
85 |
+
"eval_samples_per_second": 38.547,
|
86 |
+
"eval_steps_per_second": 4.822,
|
87 |
+
"step": 500
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.46063651591289784,
|
91 |
+
"grad_norm": 14.466177940368652,
|
92 |
+
"learning_rate": 2.7350000000000005e-06,
|
93 |
+
"loss": 0.6738,
|
94 |
+
"step": 550
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.5025125628140703,
|
98 |
+
"grad_norm": 3.282449722290039,
|
99 |
+
"learning_rate": 2.9850000000000002e-06,
|
100 |
+
"loss": 0.6517,
|
101 |
+
"step": 600
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.5443886097152428,
|
105 |
+
"grad_norm": 2.621645212173462,
|
106 |
+
"learning_rate": 3.2350000000000004e-06,
|
107 |
+
"loss": 0.6479,
|
108 |
+
"step": 650
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 0.5862646566164154,
|
112 |
+
"grad_norm": 2.4493019580841064,
|
113 |
+
"learning_rate": 3.485e-06,
|
114 |
+
"loss": 0.6321,
|
115 |
+
"step": 700
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 0.628140703517588,
|
119 |
+
"grad_norm": 1.919048547744751,
|
120 |
+
"learning_rate": 3.7350000000000002e-06,
|
121 |
+
"loss": 0.6407,
|
122 |
+
"step": 750
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.6700167504187605,
|
126 |
+
"grad_norm": 2.8260602951049805,
|
127 |
+
"learning_rate": 3.985000000000001e-06,
|
128 |
+
"loss": 0.6279,
|
129 |
+
"step": 800
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.711892797319933,
|
133 |
+
"grad_norm": 3.333841323852539,
|
134 |
+
"learning_rate": 4.235e-06,
|
135 |
+
"loss": 0.6143,
|
136 |
+
"step": 850
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.7537688442211056,
|
140 |
+
"grad_norm": 1.7883094549179077,
|
141 |
+
"learning_rate": 4.485e-06,
|
142 |
+
"loss": 0.5949,
|
143 |
+
"step": 900
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 0.7956448911222781,
|
147 |
+
"grad_norm": 2.48634934425354,
|
148 |
+
"learning_rate": 4.735e-06,
|
149 |
+
"loss": 0.5875,
|
150 |
+
"step": 950
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.8375209380234506,
|
154 |
+
"grad_norm": 5.4277873039245605,
|
155 |
+
"learning_rate": 4.9850000000000006e-06,
|
156 |
+
"loss": 0.5916,
|
157 |
+
"step": 1000
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 0.8375209380234506,
|
161 |
+
"eval_loss": 0.49684029817581177,
|
162 |
+
"eval_runtime": 219.2621,
|
163 |
+
"eval_samples_per_second": 38.716,
|
164 |
+
"eval_steps_per_second": 4.844,
|
165 |
+
"step": 1000
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.8793969849246231,
|
169 |
+
"grad_norm": 2.675995349884033,
|
170 |
+
"learning_rate": 5.235e-06,
|
171 |
+
"loss": 0.5676,
|
172 |
+
"step": 1050
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"epoch": 0.9212730318257957,
|
176 |
+
"grad_norm": 2.507474422454834,
|
177 |
+
"learning_rate": 5.485e-06,
|
178 |
+
"loss": 0.5535,
|
179 |
+
"step": 1100
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.9631490787269682,
|
183 |
+
"grad_norm": 5.360454082489014,
|
184 |
+
"learning_rate": 5.735e-06,
|
185 |
+
"loss": 0.569,
|
186 |
+
"step": 1150
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 1.0050251256281406,
|
190 |
+
"grad_norm": 2.9445533752441406,
|
191 |
+
"learning_rate": 5.985000000000001e-06,
|
192 |
+
"loss": 0.5385,
|
193 |
+
"step": 1200
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 1.0469011725293131,
|
197 |
+
"grad_norm": 2.3764073848724365,
|
198 |
+
"learning_rate": 6.235000000000001e-06,
|
199 |
+
"loss": 0.5647,
|
200 |
+
"step": 1250
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 1.0887772194304857,
|
204 |
+
"grad_norm": 4.364608287811279,
|
205 |
+
"learning_rate": 6.485000000000001e-06,
|
206 |
+
"loss": 0.5554,
|
207 |
+
"step": 1300
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 1.1306532663316582,
|
211 |
+
"grad_norm": 4.688488006591797,
|
212 |
+
"learning_rate": 6.735000000000001e-06,
|
213 |
+
"loss": 0.5676,
|
214 |
+
"step": 1350
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 1.1725293132328307,
|
218 |
+
"grad_norm": 3.0410969257354736,
|
219 |
+
"learning_rate": 6.985000000000001e-06,
|
220 |
+
"loss": 0.5374,
|
221 |
+
"step": 1400
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.2144053601340032,
|
225 |
+
"grad_norm": 6.28264856338501,
|
226 |
+
"learning_rate": 7.235000000000001e-06,
|
227 |
+
"loss": 0.5275,
|
228 |
+
"step": 1450
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 1.2562814070351758,
|
232 |
+
"grad_norm": 3.7520318031311035,
|
233 |
+
"learning_rate": 7.485000000000001e-06,
|
234 |
+
"loss": 0.5206,
|
235 |
+
"step": 1500
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 1.2562814070351758,
|
239 |
+
"eval_loss": 0.4592318534851074,
|
240 |
+
"eval_runtime": 220.2748,
|
241 |
+
"eval_samples_per_second": 38.538,
|
242 |
+
"eval_steps_per_second": 4.821,
|
243 |
+
"step": 1500
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 1.2981574539363483,
|
247 |
+
"grad_norm": 2.9252817630767822,
|
248 |
+
"learning_rate": 7.735e-06,
|
249 |
+
"loss": 0.5175,
|
250 |
+
"step": 1550
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 1.3400335008375208,
|
254 |
+
"grad_norm": 2.6018381118774414,
|
255 |
+
"learning_rate": 7.985e-06,
|
256 |
+
"loss": 0.5191,
|
257 |
+
"step": 1600
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.3819095477386933,
|
261 |
+
"grad_norm": 2.215420722961426,
|
262 |
+
"learning_rate": 8.235e-06,
|
263 |
+
"loss": 0.5274,
|
264 |
+
"step": 1650
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 1.4237855946398659,
|
268 |
+
"grad_norm": 2.787269353866577,
|
269 |
+
"learning_rate": 8.485000000000001e-06,
|
270 |
+
"loss": 0.5201,
|
271 |
+
"step": 1700
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 1.4656616415410384,
|
275 |
+
"grad_norm": 3.644702434539795,
|
276 |
+
"learning_rate": 8.735000000000002e-06,
|
277 |
+
"loss": 0.5157,
|
278 |
+
"step": 1750
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 1.507537688442211,
|
282 |
+
"grad_norm": 2.4051408767700195,
|
283 |
+
"learning_rate": 8.985000000000001e-06,
|
284 |
+
"loss": 0.5091,
|
285 |
+
"step": 1800
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 1.5494137353433834,
|
289 |
+
"grad_norm": 3.729388952255249,
|
290 |
+
"learning_rate": 9.235e-06,
|
291 |
+
"loss": 0.5134,
|
292 |
+
"step": 1850
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 1.591289782244556,
|
296 |
+
"grad_norm": 3.360603094100952,
|
297 |
+
"learning_rate": 9.485000000000002e-06,
|
298 |
+
"loss": 0.5041,
|
299 |
+
"step": 1900
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.6331658291457285,
|
303 |
+
"grad_norm": 3.6210439205169678,
|
304 |
+
"learning_rate": 9.735e-06,
|
305 |
+
"loss": 0.4993,
|
306 |
+
"step": 1950
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 1.675041876046901,
|
310 |
+
"grad_norm": 2.6245505809783936,
|
311 |
+
"learning_rate": 9.985000000000002e-06,
|
312 |
+
"loss": 0.4979,
|
313 |
+
"step": 2000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 1.675041876046901,
|
317 |
+
"eval_loss": 0.4388435482978821,
|
318 |
+
"eval_runtime": 220.5956,
|
319 |
+
"eval_samples_per_second": 38.482,
|
320 |
+
"eval_steps_per_second": 4.814,
|
321 |
+
"step": 2000
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 1.7169179229480735,
|
325 |
+
"grad_norm": 3.5152196884155273,
|
326 |
+
"learning_rate": 9.921666666666667e-06,
|
327 |
+
"loss": 0.4917,
|
328 |
+
"step": 2050
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 1.758793969849246,
|
332 |
+
"grad_norm": 2.4508583545684814,
|
333 |
+
"learning_rate": 9.838333333333334e-06,
|
334 |
+
"loss": 0.5161,
|
335 |
+
"step": 2100
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.8006700167504186,
|
339 |
+
"grad_norm": 3.44022536277771,
|
340 |
+
"learning_rate": 9.755e-06,
|
341 |
+
"loss": 0.4933,
|
342 |
+
"step": 2150
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 1.8425460636515911,
|
346 |
+
"grad_norm": 2.480679512023926,
|
347 |
+
"learning_rate": 9.671666666666668e-06,
|
348 |
+
"loss": 0.4983,
|
349 |
+
"step": 2200
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 1.8844221105527639,
|
353 |
+
"grad_norm": 2.595881462097168,
|
354 |
+
"learning_rate": 9.588333333333334e-06,
|
355 |
+
"loss": 0.4797,
|
356 |
+
"step": 2250
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 1.9262981574539364,
|
360 |
+
"grad_norm": 2.0328593254089355,
|
361 |
+
"learning_rate": 9.505000000000001e-06,
|
362 |
+
"loss": 0.483,
|
363 |
+
"step": 2300
|
364 |
+
},
|
365 |
+
{
|
366 |
+
"epoch": 1.968174204355109,
|
367 |
+
"grad_norm": 3.7075388431549072,
|
368 |
+
"learning_rate": 9.421666666666668e-06,
|
369 |
+
"loss": 0.4922,
|
370 |
+
"step": 2350
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 2.0100502512562812,
|
374 |
+
"grad_norm": 2.3989317417144775,
|
375 |
+
"learning_rate": 9.338333333333333e-06,
|
376 |
+
"loss": 0.4846,
|
377 |
+
"step": 2400
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 2.0519262981574538,
|
381 |
+
"grad_norm": 2.803882360458374,
|
382 |
+
"learning_rate": 9.255e-06,
|
383 |
+
"loss": 0.4832,
|
384 |
+
"step": 2450
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 2.0938023450586263,
|
388 |
+
"grad_norm": 3.1467437744140625,
|
389 |
+
"learning_rate": 9.171666666666667e-06,
|
390 |
+
"loss": 0.4852,
|
391 |
+
"step": 2500
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 2.0938023450586263,
|
395 |
+
"eval_loss": 0.4210895597934723,
|
396 |
+
"eval_runtime": 222.2617,
|
397 |
+
"eval_samples_per_second": 38.194,
|
398 |
+
"eval_steps_per_second": 4.778,
|
399 |
+
"step": 2500
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 2.135678391959799,
|
403 |
+
"grad_norm": 3.0684900283813477,
|
404 |
+
"learning_rate": 9.088333333333334e-06,
|
405 |
+
"loss": 0.4787,
|
406 |
+
"step": 2550
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 2.1775544388609713,
|
410 |
+
"grad_norm": 2.283045768737793,
|
411 |
+
"learning_rate": 9.005000000000001e-06,
|
412 |
+
"loss": 0.4781,
|
413 |
+
"step": 2600
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 2.219430485762144,
|
417 |
+
"grad_norm": 2.2983710765838623,
|
418 |
+
"learning_rate": 8.921666666666668e-06,
|
419 |
+
"loss": 0.4745,
|
420 |
+
"step": 2650
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 2.2613065326633164,
|
424 |
+
"grad_norm": 2.7612133026123047,
|
425 |
+
"learning_rate": 8.838333333333335e-06,
|
426 |
+
"loss": 0.4659,
|
427 |
+
"step": 2700
|
428 |
+
},
|
429 |
+
{
|
430 |
+
"epoch": 2.3031825795644894,
|
431 |
+
"grad_norm": 1.8307946920394897,
|
432 |
+
"learning_rate": 8.755e-06,
|
433 |
+
"loss": 0.4741,
|
434 |
+
"step": 2750
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 2.3450586264656614,
|
438 |
+
"grad_norm": 2.60187029838562,
|
439 |
+
"learning_rate": 8.671666666666667e-06,
|
440 |
+
"loss": 0.4603,
|
441 |
+
"step": 2800
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 2.3869346733668344,
|
445 |
+
"grad_norm": 1.789478063583374,
|
446 |
+
"learning_rate": 8.588333333333334e-06,
|
447 |
+
"loss": 0.4587,
|
448 |
+
"step": 2850
|
449 |
+
},
|
450 |
+
{
|
451 |
+
"epoch": 2.4288107202680065,
|
452 |
+
"grad_norm": 2.2422680854797363,
|
453 |
+
"learning_rate": 8.505e-06,
|
454 |
+
"loss": 0.4614,
|
455 |
+
"step": 2900
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 2.4706867671691795,
|
459 |
+
"grad_norm": 2.4102437496185303,
|
460 |
+
"learning_rate": 8.421666666666668e-06,
|
461 |
+
"loss": 0.4618,
|
462 |
+
"step": 2950
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 2.5125628140703515,
|
466 |
+
"grad_norm": 1.9738932847976685,
|
467 |
+
"learning_rate": 8.338333333333335e-06,
|
468 |
+
"loss": 0.4615,
|
469 |
+
"step": 3000
|
470 |
+
},
|
471 |
+
{
|
472 |
+
"epoch": 2.5125628140703515,
|
473 |
+
"eval_loss": 0.4087553322315216,
|
474 |
+
"eval_runtime": 221.1861,
|
475 |
+
"eval_samples_per_second": 38.379,
|
476 |
+
"eval_steps_per_second": 4.801,
|
477 |
+
"step": 3000
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 2.5544388609715245,
|
481 |
+
"grad_norm": 2.5020742416381836,
|
482 |
+
"learning_rate": 8.255000000000001e-06,
|
483 |
+
"loss": 0.4505,
|
484 |
+
"step": 3050
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 2.5963149078726966,
|
488 |
+
"grad_norm": 1.6484034061431885,
|
489 |
+
"learning_rate": 8.171666666666668e-06,
|
490 |
+
"loss": 0.4548,
|
491 |
+
"step": 3100
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 2.6381909547738696,
|
495 |
+
"grad_norm": 1.7585675716400146,
|
496 |
+
"learning_rate": 8.088333333333334e-06,
|
497 |
+
"loss": 0.4668,
|
498 |
+
"step": 3150
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 2.6800670016750416,
|
502 |
+
"grad_norm": 2.0797314643859863,
|
503 |
+
"learning_rate": 8.006666666666667e-06,
|
504 |
+
"loss": 0.4583,
|
505 |
+
"step": 3200
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 2.7219430485762146,
|
509 |
+
"grad_norm": 5.132874011993408,
|
510 |
+
"learning_rate": 7.923333333333334e-06,
|
511 |
+
"loss": 0.4573,
|
512 |
+
"step": 3250
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 2.7638190954773867,
|
516 |
+
"grad_norm": 2.005979537963867,
|
517 |
+
"learning_rate": 7.840000000000001e-06,
|
518 |
+
"loss": 0.449,
|
519 |
+
"step": 3300
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 2.8056951423785597,
|
523 |
+
"grad_norm": 2.802713394165039,
|
524 |
+
"learning_rate": 7.756666666666666e-06,
|
525 |
+
"loss": 0.448,
|
526 |
+
"step": 3350
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 2.8475711892797317,
|
530 |
+
"grad_norm": 1.8899844884872437,
|
531 |
+
"learning_rate": 7.673333333333333e-06,
|
532 |
+
"loss": 0.4558,
|
533 |
+
"step": 3400
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 2.8894472361809047,
|
537 |
+
"grad_norm": 3.7795069217681885,
|
538 |
+
"learning_rate": 7.590000000000001e-06,
|
539 |
+
"loss": 0.4533,
|
540 |
+
"step": 3450
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 2.931323283082077,
|
544 |
+
"grad_norm": 4.659352779388428,
|
545 |
+
"learning_rate": 7.506666666666668e-06,
|
546 |
+
"loss": 0.4521,
|
547 |
+
"step": 3500
|
548 |
+
},
|
549 |
+
{
|
550 |
+
"epoch": 2.931323283082077,
|
551 |
+
"eval_loss": 0.400156170129776,
|
552 |
+
"eval_runtime": 222.3025,
|
553 |
+
"eval_samples_per_second": 38.187,
|
554 |
+
"eval_steps_per_second": 4.777,
|
555 |
+
"step": 3500
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 2.9731993299832498,
|
559 |
+
"grad_norm": 2.298712730407715,
|
560 |
+
"learning_rate": 7.423333333333333e-06,
|
561 |
+
"loss": 0.4398,
|
562 |
+
"step": 3550
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 3.0150753768844223,
|
566 |
+
"grad_norm": 2.995213508605957,
|
567 |
+
"learning_rate": 7.340000000000001e-06,
|
568 |
+
"loss": 0.4496,
|
569 |
+
"step": 3600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 3.056951423785595,
|
573 |
+
"grad_norm": 2.17159366607666,
|
574 |
+
"learning_rate": 7.256666666666668e-06,
|
575 |
+
"loss": 0.4519,
|
576 |
+
"step": 3650
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 3.0988274706867673,
|
580 |
+
"grad_norm": 2.4102556705474854,
|
581 |
+
"learning_rate": 7.173333333333335e-06,
|
582 |
+
"loss": 0.4497,
|
583 |
+
"step": 3700
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 3.14070351758794,
|
587 |
+
"grad_norm": 1.7901294231414795,
|
588 |
+
"learning_rate": 7.09e-06,
|
589 |
+
"loss": 0.4545,
|
590 |
+
"step": 3750
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 3.1825795644891124,
|
594 |
+
"grad_norm": 2.869340181350708,
|
595 |
+
"learning_rate": 7.006666666666667e-06,
|
596 |
+
"loss": 0.4492,
|
597 |
+
"step": 3800
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 3.224455611390285,
|
601 |
+
"grad_norm": 2.1695075035095215,
|
602 |
+
"learning_rate": 6.9233333333333345e-06,
|
603 |
+
"loss": 0.4505,
|
604 |
+
"step": 3850
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 3.2663316582914574,
|
608 |
+
"grad_norm": 2.9931163787841797,
|
609 |
+
"learning_rate": 6.8400000000000014e-06,
|
610 |
+
"loss": 0.4421,
|
611 |
+
"step": 3900
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 3.30820770519263,
|
615 |
+
"grad_norm": 3.8682026863098145,
|
616 |
+
"learning_rate": 6.756666666666667e-06,
|
617 |
+
"loss": 0.4439,
|
618 |
+
"step": 3950
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 3.3500837520938025,
|
622 |
+
"grad_norm": 1.6634626388549805,
|
623 |
+
"learning_rate": 6.6733333333333335e-06,
|
624 |
+
"loss": 0.4431,
|
625 |
+
"step": 4000
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 3.3500837520938025,
|
629 |
+
"eval_loss": 0.39482244849205017,
|
630 |
+
"eval_runtime": 222.4171,
|
631 |
+
"eval_samples_per_second": 38.167,
|
632 |
+
"eval_steps_per_second": 4.775,
|
633 |
+
"step": 4000
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 3.391959798994975,
|
637 |
+
"grad_norm": 3.006056308746338,
|
638 |
+
"learning_rate": 6.5900000000000004e-06,
|
639 |
+
"loss": 0.4615,
|
640 |
+
"step": 4050
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 3.4338358458961475,
|
644 |
+
"grad_norm": 1.9082704782485962,
|
645 |
+
"learning_rate": 6.5066666666666665e-06,
|
646 |
+
"loss": 0.4486,
|
647 |
+
"step": 4100
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 3.47571189279732,
|
651 |
+
"grad_norm": 3.1924006938934326,
|
652 |
+
"learning_rate": 6.423333333333333e-06,
|
653 |
+
"loss": 0.4526,
|
654 |
+
"step": 4150
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 3.5175879396984926,
|
658 |
+
"grad_norm": 1.9817140102386475,
|
659 |
+
"learning_rate": 6.34e-06,
|
660 |
+
"loss": 0.4428,
|
661 |
+
"step": 4200
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 3.559463986599665,
|
665 |
+
"grad_norm": 1.3777679204940796,
|
666 |
+
"learning_rate": 6.256666666666667e-06,
|
667 |
+
"loss": 0.443,
|
668 |
+
"step": 4250
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 3.6013400335008376,
|
672 |
+
"grad_norm": 1.8882936239242554,
|
673 |
+
"learning_rate": 6.173333333333333e-06,
|
674 |
+
"loss": 0.4404,
|
675 |
+
"step": 4300
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 3.64321608040201,
|
679 |
+
"grad_norm": 1.632260799407959,
|
680 |
+
"learning_rate": 6.09e-06,
|
681 |
+
"loss": 0.4451,
|
682 |
+
"step": 4350
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 3.6850921273031827,
|
686 |
+
"grad_norm": 7.229835033416748,
|
687 |
+
"learning_rate": 6.006666666666667e-06,
|
688 |
+
"loss": 0.4352,
|
689 |
+
"step": 4400
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 3.726968174204355,
|
693 |
+
"grad_norm": 1.7340707778930664,
|
694 |
+
"learning_rate": 5.923333333333334e-06,
|
695 |
+
"loss": 0.433,
|
696 |
+
"step": 4450
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 3.7688442211055277,
|
700 |
+
"grad_norm": 1.81380295753479,
|
701 |
+
"learning_rate": 5.84e-06,
|
702 |
+
"loss": 0.4393,
|
703 |
+
"step": 4500
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 3.7688442211055277,
|
707 |
+
"eval_loss": 0.3913716673851013,
|
708 |
+
"eval_runtime": 221.2245,
|
709 |
+
"eval_samples_per_second": 38.373,
|
710 |
+
"eval_steps_per_second": 4.801,
|
711 |
+
"step": 4500
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 3.8107202680067003,
|
715 |
+
"grad_norm": 2.9950456619262695,
|
716 |
+
"learning_rate": 5.756666666666667e-06,
|
717 |
+
"loss": 0.4303,
|
718 |
+
"step": 4550
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 3.852596314907873,
|
722 |
+
"grad_norm": 1.537866473197937,
|
723 |
+
"learning_rate": 5.673333333333334e-06,
|
724 |
+
"loss": 0.4322,
|
725 |
+
"step": 4600
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 3.8944723618090453,
|
729 |
+
"grad_norm": 2.1190896034240723,
|
730 |
+
"learning_rate": 5.590000000000001e-06,
|
731 |
+
"loss": 0.4299,
|
732 |
+
"step": 4650
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 3.936348408710218,
|
736 |
+
"grad_norm": 1.5541598796844482,
|
737 |
+
"learning_rate": 5.506666666666667e-06,
|
738 |
+
"loss": 0.4259,
|
739 |
+
"step": 4700
|
740 |
+
},
|
741 |
+
{
|
742 |
+
"epoch": 3.9782244556113904,
|
743 |
+
"grad_norm": 2.0810546875,
|
744 |
+
"learning_rate": 5.423333333333334e-06,
|
745 |
+
"loss": 0.4265,
|
746 |
+
"step": 4750
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 4.0201005025125625,
|
750 |
+
"grad_norm": 1.5855557918548584,
|
751 |
+
"learning_rate": 5.3400000000000005e-06,
|
752 |
+
"loss": 0.4343,
|
753 |
+
"step": 4800
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 4.061976549413735,
|
757 |
+
"grad_norm": 2.2456538677215576,
|
758 |
+
"learning_rate": 5.256666666666667e-06,
|
759 |
+
"loss": 0.432,
|
760 |
+
"step": 4850
|
761 |
+
},
|
762 |
+
{
|
763 |
+
"epoch": 4.1038525963149075,
|
764 |
+
"grad_norm": 2.1106362342834473,
|
765 |
+
"learning_rate": 5.1733333333333335e-06,
|
766 |
+
"loss": 0.4212,
|
767 |
+
"step": 4900
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 4.1457286432160805,
|
771 |
+
"grad_norm": 2.1016457080841064,
|
772 |
+
"learning_rate": 5.09e-06,
|
773 |
+
"loss": 0.4343,
|
774 |
+
"step": 4950
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 4.187604690117253,
|
778 |
+
"grad_norm": 2.1289258003234863,
|
779 |
+
"learning_rate": 5.006666666666667e-06,
|
780 |
+
"loss": 0.4271,
|
781 |
+
"step": 5000
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 4.187604690117253,
|
785 |
+
"eval_loss": 0.3861308693885803,
|
786 |
+
"eval_runtime": 221.7311,
|
787 |
+
"eval_samples_per_second": 38.285,
|
788 |
+
"eval_steps_per_second": 4.79,
|
789 |
+
"step": 5000
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 4.2294807370184255,
|
793 |
+
"grad_norm": 1.7459477186203003,
|
794 |
+
"learning_rate": 4.923333333333334e-06,
|
795 |
+
"loss": 0.4294,
|
796 |
+
"step": 5050
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 4.271356783919598,
|
800 |
+
"grad_norm": 2.2993977069854736,
|
801 |
+
"learning_rate": 4.84e-06,
|
802 |
+
"loss": 0.4305,
|
803 |
+
"step": 5100
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 4.313232830820771,
|
807 |
+
"grad_norm": 2.3953640460968018,
|
808 |
+
"learning_rate": 4.756666666666667e-06,
|
809 |
+
"loss": 0.4297,
|
810 |
+
"step": 5150
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 4.355108877721943,
|
814 |
+
"grad_norm": 1.3711094856262207,
|
815 |
+
"learning_rate": 4.673333333333333e-06,
|
816 |
+
"loss": 0.4277,
|
817 |
+
"step": 5200
|
818 |
+
},
|
819 |
+
{
|
820 |
+
"epoch": 4.396984924623116,
|
821 |
+
"grad_norm": 1.373367428779602,
|
822 |
+
"learning_rate": 4.590000000000001e-06,
|
823 |
+
"loss": 0.424,
|
824 |
+
"step": 5250
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 4.438860971524288,
|
828 |
+
"grad_norm": 2.1267800331115723,
|
829 |
+
"learning_rate": 4.506666666666667e-06,
|
830 |
+
"loss": 0.4244,
|
831 |
+
"step": 5300
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 4.480737018425461,
|
835 |
+
"grad_norm": 1.67445707321167,
|
836 |
+
"learning_rate": 4.423333333333334e-06,
|
837 |
+
"loss": 0.4246,
|
838 |
+
"step": 5350
|
839 |
+
},
|
840 |
+
{
|
841 |
+
"epoch": 4.522613065326633,
|
842 |
+
"grad_norm": 2.8858883380889893,
|
843 |
+
"learning_rate": 4.34e-06,
|
844 |
+
"loss": 0.4363,
|
845 |
+
"step": 5400
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 4.564489112227806,
|
849 |
+
"grad_norm": 1.7339367866516113,
|
850 |
+
"learning_rate": 4.256666666666668e-06,
|
851 |
+
"loss": 0.4257,
|
852 |
+
"step": 5450
|
853 |
+
},
|
854 |
+
{
|
855 |
+
"epoch": 4.606365159128979,
|
856 |
+
"grad_norm": 1.892198085784912,
|
857 |
+
"learning_rate": 4.173333333333334e-06,
|
858 |
+
"loss": 0.4317,
|
859 |
+
"step": 5500
|
860 |
+
},
|
861 |
+
{
|
862 |
+
"epoch": 4.606365159128979,
|
863 |
+
"eval_loss": 0.3835948705673218,
|
864 |
+
"eval_runtime": 220.3794,
|
865 |
+
"eval_samples_per_second": 38.52,
|
866 |
+
"eval_steps_per_second": 4.819,
|
867 |
+
"step": 5500
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 4.648241206030151,
|
871 |
+
"grad_norm": 1.3021107912063599,
|
872 |
+
"learning_rate": 4.09e-06,
|
873 |
+
"loss": 0.4239,
|
874 |
+
"step": 5550
|
875 |
+
},
|
876 |
+
{
|
877 |
+
"epoch": 4.690117252931323,
|
878 |
+
"grad_norm": 1.9206578731536865,
|
879 |
+
"learning_rate": 4.006666666666667e-06,
|
880 |
+
"loss": 0.4296,
|
881 |
+
"step": 5600
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 4.731993299832496,
|
885 |
+
"grad_norm": 1.5515390634536743,
|
886 |
+
"learning_rate": 3.923333333333334e-06,
|
887 |
+
"loss": 0.4253,
|
888 |
+
"step": 5650
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 4.773869346733669,
|
892 |
+
"grad_norm": 1.9020607471466064,
|
893 |
+
"learning_rate": 3.8400000000000005e-06,
|
894 |
+
"loss": 0.4232,
|
895 |
+
"step": 5700
|
896 |
+
},
|
897 |
+
{
|
898 |
+
"epoch": 4.815745393634841,
|
899 |
+
"grad_norm": 1.831719160079956,
|
900 |
+
"learning_rate": 3.756666666666667e-06,
|
901 |
+
"loss": 0.4264,
|
902 |
+
"step": 5750
|
903 |
+
},
|
904 |
+
{
|
905 |
+
"epoch": 4.857621440536013,
|
906 |
+
"grad_norm": 1.7857838869094849,
|
907 |
+
"learning_rate": 3.673333333333334e-06,
|
908 |
+
"loss": 0.4217,
|
909 |
+
"step": 5800
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 4.899497487437186,
|
913 |
+
"grad_norm": 1.8334425687789917,
|
914 |
+
"learning_rate": 3.5900000000000004e-06,
|
915 |
+
"loss": 0.4256,
|
916 |
+
"step": 5850
|
917 |
+
},
|
918 |
+
{
|
919 |
+
"epoch": 4.941373534338359,
|
920 |
+
"grad_norm": 1.8758697509765625,
|
921 |
+
"learning_rate": 3.5066666666666673e-06,
|
922 |
+
"loss": 0.4238,
|
923 |
+
"step": 5900
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 4.983249581239531,
|
927 |
+
"grad_norm": 2.417583465576172,
|
928 |
+
"learning_rate": 3.4233333333333333e-06,
|
929 |
+
"loss": 0.4194,
|
930 |
+
"step": 5950
|
931 |
+
},
|
932 |
+
{
|
933 |
+
"epoch": 5.025125628140704,
|
934 |
+
"grad_norm": 1.6724933385849,
|
935 |
+
"learning_rate": 3.3400000000000006e-06,
|
936 |
+
"loss": 0.4265,
|
937 |
+
"step": 6000
|
938 |
+
},
|
939 |
+
{
|
940 |
+
"epoch": 5.025125628140704,
|
941 |
+
"eval_loss": 0.38090234994888306,
|
942 |
+
"eval_runtime": 221.6516,
|
943 |
+
"eval_samples_per_second": 38.299,
|
944 |
+
"eval_steps_per_second": 4.791,
|
945 |
+
"step": 6000
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 5.067001675041876,
|
949 |
+
"grad_norm": 1.7485418319702148,
|
950 |
+
"learning_rate": 3.2566666666666667e-06,
|
951 |
+
"loss": 0.4154,
|
952 |
+
"step": 6050
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 5.108877721943049,
|
956 |
+
"grad_norm": 1.8903107643127441,
|
957 |
+
"learning_rate": 3.173333333333334e-06,
|
958 |
+
"loss": 0.4224,
|
959 |
+
"step": 6100
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 5.150753768844221,
|
963 |
+
"grad_norm": 1.360568881034851,
|
964 |
+
"learning_rate": 3.09e-06,
|
965 |
+
"loss": 0.4207,
|
966 |
+
"step": 6150
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 5.192629815745394,
|
970 |
+
"grad_norm": 1.8849430084228516,
|
971 |
+
"learning_rate": 3.0066666666666674e-06,
|
972 |
+
"loss": 0.4193,
|
973 |
+
"step": 6200
|
974 |
+
},
|
975 |
+
{
|
976 |
+
"epoch": 5.234505862646566,
|
977 |
+
"grad_norm": 1.2618883848190308,
|
978 |
+
"learning_rate": 2.9233333333333334e-06,
|
979 |
+
"loss": 0.4174,
|
980 |
+
"step": 6250
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 5.276381909547739,
|
984 |
+
"grad_norm": 1.923030972480774,
|
985 |
+
"learning_rate": 2.84e-06,
|
986 |
+
"loss": 0.4188,
|
987 |
+
"step": 6300
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 5.318257956448911,
|
991 |
+
"grad_norm": 1.4617573022842407,
|
992 |
+
"learning_rate": 2.756666666666667e-06,
|
993 |
+
"loss": 0.4195,
|
994 |
+
"step": 6350
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 5.360134003350084,
|
998 |
+
"grad_norm": 1.308645248413086,
|
999 |
+
"learning_rate": 2.6733333333333333e-06,
|
1000 |
+
"loss": 0.4265,
|
1001 |
+
"step": 6400
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 5.402010050251256,
|
1005 |
+
"grad_norm": 1.4184455871582031,
|
1006 |
+
"learning_rate": 2.59e-06,
|
1007 |
+
"loss": 0.4202,
|
1008 |
+
"step": 6450
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 5.443886097152429,
|
1012 |
+
"grad_norm": 1.2012479305267334,
|
1013 |
+
"learning_rate": 2.5066666666666667e-06,
|
1014 |
+
"loss": 0.424,
|
1015 |
+
"step": 6500
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 5.443886097152429,
|
1019 |
+
"eval_loss": 0.3793572783470154,
|
1020 |
+
"eval_runtime": 221.7764,
|
1021 |
+
"eval_samples_per_second": 38.277,
|
1022 |
+
"eval_steps_per_second": 4.789,
|
1023 |
+
"step": 6500
|
1024 |
+
},
|
1025 |
+
{
|
1026 |
+
"epoch": 5.485762144053601,
|
1027 |
+
"grad_norm": 1.3103199005126953,
|
1028 |
+
"learning_rate": 2.4233333333333336e-06,
|
1029 |
+
"loss": 0.4179,
|
1030 |
+
"step": 6550
|
1031 |
+
},
|
1032 |
+
{
|
1033 |
+
"epoch": 5.527638190954773,
|
1034 |
+
"grad_norm": 1.4100350141525269,
|
1035 |
+
"learning_rate": 2.3400000000000005e-06,
|
1036 |
+
"loss": 0.4201,
|
1037 |
+
"step": 6600
|
1038 |
+
},
|
1039 |
+
{
|
1040 |
+
"epoch": 5.569514237855946,
|
1041 |
+
"grad_norm": 1.944542646408081,
|
1042 |
+
"learning_rate": 2.2566666666666665e-06,
|
1043 |
+
"loss": 0.4157,
|
1044 |
+
"step": 6650
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 5.611390284757119,
|
1048 |
+
"grad_norm": 1.759690284729004,
|
1049 |
+
"learning_rate": 2.1733333333333334e-06,
|
1050 |
+
"loss": 0.4135,
|
1051 |
+
"step": 6700
|
1052 |
+
},
|
1053 |
+
{
|
1054 |
+
"epoch": 5.653266331658291,
|
1055 |
+
"grad_norm": 1.6196409463882446,
|
1056 |
+
"learning_rate": 2.09e-06,
|
1057 |
+
"loss": 0.4218,
|
1058 |
+
"step": 6750
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 5.695142378559464,
|
1062 |
+
"grad_norm": 1.4138773679733276,
|
1063 |
+
"learning_rate": 2.006666666666667e-06,
|
1064 |
+
"loss": 0.4228,
|
1065 |
+
"step": 6800
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 5.7370184254606365,
|
1069 |
+
"grad_norm": 1.7974828481674194,
|
1070 |
+
"learning_rate": 1.9233333333333333e-06,
|
1071 |
+
"loss": 0.4242,
|
1072 |
+
"step": 6850
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 5.778894472361809,
|
1076 |
+
"grad_norm": 1.697729229927063,
|
1077 |
+
"learning_rate": 1.8400000000000002e-06,
|
1078 |
+
"loss": 0.4133,
|
1079 |
+
"step": 6900
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 5.8207705192629815,
|
1083 |
+
"grad_norm": 1.2323369979858398,
|
1084 |
+
"learning_rate": 1.7566666666666669e-06,
|
1085 |
+
"loss": 0.4205,
|
1086 |
+
"step": 6950
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 5.8626465661641545,
|
1090 |
+
"grad_norm": 1.0918774604797363,
|
1091 |
+
"learning_rate": 1.6733333333333335e-06,
|
1092 |
+
"loss": 0.4123,
|
1093 |
+
"step": 7000
|
1094 |
+
},
|
1095 |
+
{
|
1096 |
+
"epoch": 5.8626465661641545,
|
1097 |
+
"eval_loss": 0.3785873353481293,
|
1098 |
+
"eval_runtime": 222.4775,
|
1099 |
+
"eval_samples_per_second": 38.157,
|
1100 |
+
"eval_steps_per_second": 4.774,
|
1101 |
+
"step": 7000
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 5.9045226130653266,
|
1105 |
+
"grad_norm": 1.4229035377502441,
|
1106 |
+
"learning_rate": 1.5900000000000002e-06,
|
1107 |
+
"loss": 0.4263,
|
1108 |
+
"step": 7050
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 5.9463986599664995,
|
1112 |
+
"grad_norm": 1.8028751611709595,
|
1113 |
+
"learning_rate": 1.506666666666667e-06,
|
1114 |
+
"loss": 0.4124,
|
1115 |
+
"step": 7100
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 5.988274706867672,
|
1119 |
+
"grad_norm": 1.8243662118911743,
|
1120 |
+
"learning_rate": 1.4233333333333336e-06,
|
1121 |
+
"loss": 0.4212,
|
1122 |
+
"step": 7150
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 6.030150753768845,
|
1126 |
+
"grad_norm": 1.7111196517944336,
|
1127 |
+
"learning_rate": 1.34e-06,
|
1128 |
+
"loss": 0.4107,
|
1129 |
+
"step": 7200
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 6.072026800670017,
|
1133 |
+
"grad_norm": 1.0116655826568604,
|
1134 |
+
"learning_rate": 1.2566666666666668e-06,
|
1135 |
+
"loss": 0.4217,
|
1136 |
+
"step": 7250
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 6.11390284757119,
|
1140 |
+
"grad_norm": 1.32182776927948,
|
1141 |
+
"learning_rate": 1.1733333333333335e-06,
|
1142 |
+
"loss": 0.4124,
|
1143 |
+
"step": 7300
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 6.155778894472362,
|
1147 |
+
"grad_norm": 1.2914599180221558,
|
1148 |
+
"learning_rate": 1.0900000000000002e-06,
|
1149 |
+
"loss": 0.4208,
|
1150 |
+
"step": 7350
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 6.197654941373535,
|
1154 |
+
"grad_norm": 1.106825351715088,
|
1155 |
+
"learning_rate": 1.0066666666666668e-06,
|
1156 |
+
"loss": 0.4128,
|
1157 |
+
"step": 7400
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 6.239530988274707,
|
1161 |
+
"grad_norm": 1.452335000038147,
|
1162 |
+
"learning_rate": 9.233333333333334e-07,
|
1163 |
+
"loss": 0.4078,
|
1164 |
+
"step": 7450
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 6.28140703517588,
|
1168 |
+
"grad_norm": 0.9933892488479614,
|
1169 |
+
"learning_rate": 8.400000000000001e-07,
|
1170 |
+
"loss": 0.4117,
|
1171 |
+
"step": 7500
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 6.28140703517588,
|
1175 |
+
"eval_loss": 0.3775756061077118,
|
1176 |
+
"eval_runtime": 222.6181,
|
1177 |
+
"eval_samples_per_second": 38.133,
|
1178 |
+
"eval_steps_per_second": 4.771,
|
1179 |
+
"step": 7500
|
1180 |
+
}
|
1181 |
+
],
|
1182 |
+
"logging_steps": 50,
|
1183 |
+
"max_steps": 8000,
|
1184 |
+
"num_input_tokens_seen": 0,
|
1185 |
+
"num_train_epochs": 7,
|
1186 |
+
"save_steps": 500,
|
1187 |
+
"stateful_callbacks": {
|
1188 |
+
"TrainerControl": {
|
1189 |
+
"args": {
|
1190 |
+
"should_epoch_stop": false,
|
1191 |
+
"should_evaluate": false,
|
1192 |
+
"should_log": false,
|
1193 |
+
"should_save": true,
|
1194 |
+
"should_training_stop": false
|
1195 |
+
},
|
1196 |
+
"attributes": {}
|
1197 |
+
}
|
1198 |
+
},
|
1199 |
+
"total_flos": 7.276067271254074e+16,
|
1200 |
+
"train_batch_size": 32,
|
1201 |
+
"trial_name": null,
|
1202 |
+
"trial_params": null
|
1203 |
+
}
|
checkpoint-7500/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6248ade84a48a52ce5a07ff6e203bd335b96af70b5ce5ebd92ad797d93be969c
|
3 |
+
size 5304
|