Upload checkpoint-5204
Browse files- checkpoint-5204/config.json +152 -0
- checkpoint-5204/generation_config.json +241 -0
- checkpoint-5204/model.safetensors +3 -0
- checkpoint-5204/optimizer.pt +3 -0
- checkpoint-5204/preprocessor_config.json +14 -0
- checkpoint-5204/rng_state_0.pth +3 -0
- checkpoint-5204/rng_state_1.pth +3 -0
- checkpoint-5204/rng_state_2.pth +3 -0
- checkpoint-5204/rng_state_3.pth +3 -0
- checkpoint-5204/rng_state_4.pth +3 -0
- checkpoint-5204/rng_state_5.pth +3 -0
- checkpoint-5204/rng_state_6.pth +3 -0
- checkpoint-5204/rng_state_7.pth +3 -0
- checkpoint-5204/scheduler.pt +3 -0
- checkpoint-5204/trainer_state.json +762 -0
- checkpoint-5204/training_args.bin +3 -0
checkpoint-5204/config.json
ADDED
|
@@ -0,0 +1,152 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_dropout": 0.0,
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"apply_spec_augment": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"WhisperForConditionalGeneration"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"begin_suppress_tokens": [
|
| 10 |
+
220,
|
| 11 |
+
50257
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"d_model": 1024,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 4096,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 24,
|
| 20 |
+
"decoder_start_token_id": 50258,
|
| 21 |
+
"dropout": 0.0,
|
| 22 |
+
"dtype": "float32",
|
| 23 |
+
"encoder_attention_heads": 16,
|
| 24 |
+
"encoder_ffn_dim": 4096,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 24,
|
| 27 |
+
"eos_token_id": 50257,
|
| 28 |
+
"forced_decoder_ids": [
|
| 29 |
+
[
|
| 30 |
+
1,
|
| 31 |
+
50259
|
| 32 |
+
],
|
| 33 |
+
[
|
| 34 |
+
2,
|
| 35 |
+
50359
|
| 36 |
+
],
|
| 37 |
+
[
|
| 38 |
+
3,
|
| 39 |
+
50363
|
| 40 |
+
]
|
| 41 |
+
],
|
| 42 |
+
"init_std": 0.02,
|
| 43 |
+
"is_encoder_decoder": true,
|
| 44 |
+
"mask_feature_length": 10,
|
| 45 |
+
"mask_feature_min_masks": 0,
|
| 46 |
+
"mask_feature_prob": 0.0,
|
| 47 |
+
"mask_time_length": 10,
|
| 48 |
+
"mask_time_min_masks": 2,
|
| 49 |
+
"mask_time_prob": 0.05,
|
| 50 |
+
"max_source_positions": 1500,
|
| 51 |
+
"max_target_positions": 448,
|
| 52 |
+
"median_filter_width": 7,
|
| 53 |
+
"model_type": "whisper",
|
| 54 |
+
"num_mel_bins": 80,
|
| 55 |
+
"pad_token_id": 50257,
|
| 56 |
+
"scale_embedding": false,
|
| 57 |
+
"suppress_tokens": [
|
| 58 |
+
1,
|
| 59 |
+
2,
|
| 60 |
+
7,
|
| 61 |
+
8,
|
| 62 |
+
9,
|
| 63 |
+
10,
|
| 64 |
+
14,
|
| 65 |
+
25,
|
| 66 |
+
26,
|
| 67 |
+
27,
|
| 68 |
+
28,
|
| 69 |
+
29,
|
| 70 |
+
31,
|
| 71 |
+
58,
|
| 72 |
+
59,
|
| 73 |
+
60,
|
| 74 |
+
61,
|
| 75 |
+
62,
|
| 76 |
+
63,
|
| 77 |
+
90,
|
| 78 |
+
91,
|
| 79 |
+
92,
|
| 80 |
+
93,
|
| 81 |
+
359,
|
| 82 |
+
503,
|
| 83 |
+
522,
|
| 84 |
+
542,
|
| 85 |
+
873,
|
| 86 |
+
893,
|
| 87 |
+
902,
|
| 88 |
+
918,
|
| 89 |
+
922,
|
| 90 |
+
931,
|
| 91 |
+
1350,
|
| 92 |
+
1853,
|
| 93 |
+
1982,
|
| 94 |
+
2460,
|
| 95 |
+
2627,
|
| 96 |
+
3246,
|
| 97 |
+
3253,
|
| 98 |
+
3268,
|
| 99 |
+
3536,
|
| 100 |
+
3846,
|
| 101 |
+
3961,
|
| 102 |
+
4183,
|
| 103 |
+
4667,
|
| 104 |
+
6585,
|
| 105 |
+
6647,
|
| 106 |
+
7273,
|
| 107 |
+
9061,
|
| 108 |
+
9383,
|
| 109 |
+
10428,
|
| 110 |
+
10929,
|
| 111 |
+
11938,
|
| 112 |
+
12033,
|
| 113 |
+
12331,
|
| 114 |
+
12562,
|
| 115 |
+
13793,
|
| 116 |
+
14157,
|
| 117 |
+
14635,
|
| 118 |
+
15265,
|
| 119 |
+
15618,
|
| 120 |
+
16553,
|
| 121 |
+
16604,
|
| 122 |
+
18362,
|
| 123 |
+
18956,
|
| 124 |
+
20075,
|
| 125 |
+
21675,
|
| 126 |
+
22520,
|
| 127 |
+
26130,
|
| 128 |
+
26161,
|
| 129 |
+
26435,
|
| 130 |
+
28279,
|
| 131 |
+
29464,
|
| 132 |
+
31650,
|
| 133 |
+
32302,
|
| 134 |
+
32470,
|
| 135 |
+
36865,
|
| 136 |
+
42863,
|
| 137 |
+
47425,
|
| 138 |
+
49870,
|
| 139 |
+
50254,
|
| 140 |
+
50258,
|
| 141 |
+
50358,
|
| 142 |
+
50359,
|
| 143 |
+
50360,
|
| 144 |
+
50361,
|
| 145 |
+
50362
|
| 146 |
+
],
|
| 147 |
+
"tie_word_embeddings": true,
|
| 148 |
+
"transformers_version": "5.4.0",
|
| 149 |
+
"use_cache": false,
|
| 150 |
+
"use_weighted_layer_sum": false,
|
| 151 |
+
"vocab_size": 51865
|
| 152 |
+
}
|
checkpoint-5204/generation_config.json
ADDED
|
@@ -0,0 +1,241 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alignment_heads": [
|
| 3 |
+
[
|
| 4 |
+
13,
|
| 5 |
+
15
|
| 6 |
+
],
|
| 7 |
+
[
|
| 8 |
+
15,
|
| 9 |
+
4
|
| 10 |
+
],
|
| 11 |
+
[
|
| 12 |
+
15,
|
| 13 |
+
15
|
| 14 |
+
],
|
| 15 |
+
[
|
| 16 |
+
16,
|
| 17 |
+
1
|
| 18 |
+
],
|
| 19 |
+
[
|
| 20 |
+
20,
|
| 21 |
+
0
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
23,
|
| 25 |
+
4
|
| 26 |
+
]
|
| 27 |
+
],
|
| 28 |
+
"begin_suppress_tokens": [
|
| 29 |
+
220,
|
| 30 |
+
50257
|
| 31 |
+
],
|
| 32 |
+
"bos_token_id": 50257,
|
| 33 |
+
"decoder_start_token_id": 50258,
|
| 34 |
+
"eos_token_id": 50257,
|
| 35 |
+
"forced_decoder_ids": null,
|
| 36 |
+
"is_multilingual": true,
|
| 37 |
+
"lang_to_id": {
|
| 38 |
+
"<|af|>": 50327,
|
| 39 |
+
"<|am|>": 50334,
|
| 40 |
+
"<|ar|>": 50272,
|
| 41 |
+
"<|as|>": 50350,
|
| 42 |
+
"<|az|>": 50304,
|
| 43 |
+
"<|ba|>": 50355,
|
| 44 |
+
"<|be|>": 50330,
|
| 45 |
+
"<|bg|>": 50292,
|
| 46 |
+
"<|bn|>": 50302,
|
| 47 |
+
"<|bo|>": 50347,
|
| 48 |
+
"<|br|>": 50309,
|
| 49 |
+
"<|bs|>": 50315,
|
| 50 |
+
"<|ca|>": 50270,
|
| 51 |
+
"<|cs|>": 50283,
|
| 52 |
+
"<|cy|>": 50297,
|
| 53 |
+
"<|da|>": 50285,
|
| 54 |
+
"<|de|>": 50261,
|
| 55 |
+
"<|el|>": 50281,
|
| 56 |
+
"<|en|>": 50259,
|
| 57 |
+
"<|es|>": 50262,
|
| 58 |
+
"<|et|>": 50307,
|
| 59 |
+
"<|eu|>": 50310,
|
| 60 |
+
"<|fa|>": 50300,
|
| 61 |
+
"<|fi|>": 50277,
|
| 62 |
+
"<|fo|>": 50338,
|
| 63 |
+
"<|fr|>": 50265,
|
| 64 |
+
"<|gl|>": 50319,
|
| 65 |
+
"<|gu|>": 50333,
|
| 66 |
+
"<|haw|>": 50352,
|
| 67 |
+
"<|ha|>": 50354,
|
| 68 |
+
"<|he|>": 50279,
|
| 69 |
+
"<|hi|>": 50276,
|
| 70 |
+
"<|hr|>": 50291,
|
| 71 |
+
"<|ht|>": 50339,
|
| 72 |
+
"<|hu|>": 50286,
|
| 73 |
+
"<|hy|>": 50312,
|
| 74 |
+
"<|id|>": 50275,
|
| 75 |
+
"<|is|>": 50311,
|
| 76 |
+
"<|it|>": 50274,
|
| 77 |
+
"<|ja|>": 50266,
|
| 78 |
+
"<|jw|>": 50356,
|
| 79 |
+
"<|ka|>": 50329,
|
| 80 |
+
"<|kk|>": 50316,
|
| 81 |
+
"<|km|>": 50323,
|
| 82 |
+
"<|kn|>": 50306,
|
| 83 |
+
"<|ko|>": 50264,
|
| 84 |
+
"<|la|>": 50294,
|
| 85 |
+
"<|lb|>": 50345,
|
| 86 |
+
"<|ln|>": 50353,
|
| 87 |
+
"<|lo|>": 50336,
|
| 88 |
+
"<|lt|>": 50293,
|
| 89 |
+
"<|lv|>": 50301,
|
| 90 |
+
"<|mg|>": 50349,
|
| 91 |
+
"<|mi|>": 50295,
|
| 92 |
+
"<|mk|>": 50308,
|
| 93 |
+
"<|ml|>": 50296,
|
| 94 |
+
"<|mn|>": 50314,
|
| 95 |
+
"<|mr|>": 50320,
|
| 96 |
+
"<|ms|>": 50282,
|
| 97 |
+
"<|mt|>": 50343,
|
| 98 |
+
"<|my|>": 50346,
|
| 99 |
+
"<|ne|>": 50313,
|
| 100 |
+
"<|nl|>": 50271,
|
| 101 |
+
"<|nn|>": 50342,
|
| 102 |
+
"<|no|>": 50288,
|
| 103 |
+
"<|oc|>": 50328,
|
| 104 |
+
"<|pa|>": 50321,
|
| 105 |
+
"<|pl|>": 50269,
|
| 106 |
+
"<|ps|>": 50340,
|
| 107 |
+
"<|pt|>": 50267,
|
| 108 |
+
"<|ro|>": 50284,
|
| 109 |
+
"<|ru|>": 50263,
|
| 110 |
+
"<|sa|>": 50344,
|
| 111 |
+
"<|sd|>": 50332,
|
| 112 |
+
"<|si|>": 50322,
|
| 113 |
+
"<|sk|>": 50298,
|
| 114 |
+
"<|sl|>": 50305,
|
| 115 |
+
"<|sn|>": 50324,
|
| 116 |
+
"<|so|>": 50326,
|
| 117 |
+
"<|sq|>": 50317,
|
| 118 |
+
"<|sr|>": 50303,
|
| 119 |
+
"<|su|>": 50357,
|
| 120 |
+
"<|sv|>": 50273,
|
| 121 |
+
"<|sw|>": 50318,
|
| 122 |
+
"<|ta|>": 50287,
|
| 123 |
+
"<|te|>": 50299,
|
| 124 |
+
"<|tg|>": 50331,
|
| 125 |
+
"<|th|>": 50289,
|
| 126 |
+
"<|tk|>": 50341,
|
| 127 |
+
"<|tl|>": 50348,
|
| 128 |
+
"<|tr|>": 50268,
|
| 129 |
+
"<|tt|>": 50351,
|
| 130 |
+
"<|uk|>": 50280,
|
| 131 |
+
"<|ur|>": 50290,
|
| 132 |
+
"<|uz|>": 50337,
|
| 133 |
+
"<|vi|>": 50278,
|
| 134 |
+
"<|yi|>": 50335,
|
| 135 |
+
"<|yo|>": 50325,
|
| 136 |
+
"<|zh|>": 50260
|
| 137 |
+
},
|
| 138 |
+
"language": "Hindi",
|
| 139 |
+
"max_initial_timestamp_index": 50,
|
| 140 |
+
"max_length": 448,
|
| 141 |
+
"no_timestamps_token_id": 50363,
|
| 142 |
+
"pad_token_id": 50257,
|
| 143 |
+
"prev_sot_token_id": 50361,
|
| 144 |
+
"return_timestamps": false,
|
| 145 |
+
"suppress_tokens": [
|
| 146 |
+
1,
|
| 147 |
+
2,
|
| 148 |
+
7,
|
| 149 |
+
8,
|
| 150 |
+
9,
|
| 151 |
+
10,
|
| 152 |
+
14,
|
| 153 |
+
25,
|
| 154 |
+
26,
|
| 155 |
+
27,
|
| 156 |
+
28,
|
| 157 |
+
29,
|
| 158 |
+
31,
|
| 159 |
+
58,
|
| 160 |
+
59,
|
| 161 |
+
60,
|
| 162 |
+
61,
|
| 163 |
+
62,
|
| 164 |
+
63,
|
| 165 |
+
90,
|
| 166 |
+
91,
|
| 167 |
+
92,
|
| 168 |
+
93,
|
| 169 |
+
359,
|
| 170 |
+
503,
|
| 171 |
+
522,
|
| 172 |
+
542,
|
| 173 |
+
873,
|
| 174 |
+
893,
|
| 175 |
+
902,
|
| 176 |
+
918,
|
| 177 |
+
922,
|
| 178 |
+
931,
|
| 179 |
+
1350,
|
| 180 |
+
1853,
|
| 181 |
+
1982,
|
| 182 |
+
2460,
|
| 183 |
+
2627,
|
| 184 |
+
3246,
|
| 185 |
+
3253,
|
| 186 |
+
3268,
|
| 187 |
+
3536,
|
| 188 |
+
3846,
|
| 189 |
+
3961,
|
| 190 |
+
4183,
|
| 191 |
+
4667,
|
| 192 |
+
6585,
|
| 193 |
+
6647,
|
| 194 |
+
7273,
|
| 195 |
+
9061,
|
| 196 |
+
9383,
|
| 197 |
+
10428,
|
| 198 |
+
10929,
|
| 199 |
+
11938,
|
| 200 |
+
12033,
|
| 201 |
+
12331,
|
| 202 |
+
12562,
|
| 203 |
+
13793,
|
| 204 |
+
14157,
|
| 205 |
+
14635,
|
| 206 |
+
15265,
|
| 207 |
+
15618,
|
| 208 |
+
16553,
|
| 209 |
+
16604,
|
| 210 |
+
18362,
|
| 211 |
+
18956,
|
| 212 |
+
20075,
|
| 213 |
+
21675,
|
| 214 |
+
22520,
|
| 215 |
+
26130,
|
| 216 |
+
26161,
|
| 217 |
+
26435,
|
| 218 |
+
28279,
|
| 219 |
+
29464,
|
| 220 |
+
31650,
|
| 221 |
+
32302,
|
| 222 |
+
32470,
|
| 223 |
+
36865,
|
| 224 |
+
42863,
|
| 225 |
+
47425,
|
| 226 |
+
49870,
|
| 227 |
+
50254,
|
| 228 |
+
50258,
|
| 229 |
+
50358,
|
| 230 |
+
50359,
|
| 231 |
+
50360,
|
| 232 |
+
50361,
|
| 233 |
+
50362
|
| 234 |
+
],
|
| 235 |
+
"task": "transcribe",
|
| 236 |
+
"task_to_id": {
|
| 237 |
+
"transcribe": 50359,
|
| 238 |
+
"translate": 50358
|
| 239 |
+
},
|
| 240 |
+
"transformers_version": "5.4.0"
|
| 241 |
+
}
|
checkpoint-5204/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2064e2d659a7aa86f46a5cf95f708036575009e7564ba5c942ec9cdbac2c2685
|
| 3 |
+
size 3055544304
|
checkpoint-5204/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e6419e9e2a3e889e19a20ad4bda0bf27fc5e4e9ac6e6b59c3223a271057c966
|
| 3 |
+
size 6111697804
|
checkpoint-5204/preprocessor_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"dither": 0.0,
|
| 4 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 5 |
+
"feature_size": 80,
|
| 6 |
+
"hop_length": 160,
|
| 7 |
+
"n_fft": 400,
|
| 8 |
+
"n_samples": 480000,
|
| 9 |
+
"nb_max_frames": 3000,
|
| 10 |
+
"padding_side": "right",
|
| 11 |
+
"padding_value": 0.0,
|
| 12 |
+
"return_attention_mask": false,
|
| 13 |
+
"sampling_rate": 16000
|
| 14 |
+
}
|
checkpoint-5204/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e149a2d21d702cce0057ef8b9bf57d98cae0103d612e4dcbd2a747d419459d73
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32a8d163e56347dc738ff9f35bb2ead5c1c750ddbbb31d9b86c4255bfcfc047b
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0bf74d96af0ae5eb1999b52f5543b933096d44a026bf730d17185f4faffbecd1
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bcf9da40c0c8749ae9c2bd1c9bae40db707617cffc7dc157a1de9a4da64d831
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:942a5e22bcb767f40850e9a9bd1112b67f14538ce71f09a74d2c86de6e917f5f
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94f15f813053adaab0cb48567158c59b31df35b068cf02ec7074cefea2afaf4a
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5bfaf45e99ef15a988af4d90fc8dc3fb6a190a68f7963d24caa63a83fd77505
|
| 3 |
+
size 16389
|
checkpoint-5204/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea9edd6835a74f1f16de5a3021bead3d6c205a20f019b590ae3d64bcd55a71e3
|
| 3 |
+
size 16389
|
checkpoint-5204/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a4a0b413b8a606f6c96afdf44db8248aaa54ba1890b5ec473ac4d42edf67cb9
|
| 3 |
+
size 1465
|
checkpoint-5204/trainer_state.json
ADDED
|
@@ -0,0 +1,762 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 4.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 5204,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.03843197540353574,
|
| 14 |
+
"grad_norm": 24.96473503112793,
|
| 15 |
+
"learning_rate": 7.000000000000001e-07,
|
| 16 |
+
"loss": 6.548916015625,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.07686395080707148,
|
| 21 |
+
"grad_norm": 15.567505836486816,
|
| 22 |
+
"learning_rate": 1.4142857142857144e-06,
|
| 23 |
+
"loss": 4.4738397216796875,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.11529592621060722,
|
| 28 |
+
"grad_norm": 13.502691268920898,
|
| 29 |
+
"learning_rate": 2.1285714285714287e-06,
|
| 30 |
+
"loss": 3.2899395751953127,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.15372790161414296,
|
| 35 |
+
"grad_norm": 9.215691566467285,
|
| 36 |
+
"learning_rate": 2.8428571428571432e-06,
|
| 37 |
+
"loss": 2.035605621337891,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.1921598770176787,
|
| 42 |
+
"grad_norm": 9.399081230163574,
|
| 43 |
+
"learning_rate": 3.5571428571428573e-06,
|
| 44 |
+
"loss": 1.7253753662109375,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.23059185242121444,
|
| 49 |
+
"grad_norm": 9.485953330993652,
|
| 50 |
+
"learning_rate": 4.271428571428572e-06,
|
| 51 |
+
"loss": 1.5824163818359376,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.2690238278247502,
|
| 56 |
+
"grad_norm": 8.881640434265137,
|
| 57 |
+
"learning_rate": 4.9857142857142855e-06,
|
| 58 |
+
"loss": 1.4994566345214844,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.3074558032282859,
|
| 63 |
+
"grad_norm": 9.438455581665039,
|
| 64 |
+
"learning_rate": 5.7e-06,
|
| 65 |
+
"loss": 1.4217744445800782,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.3458877786318217,
|
| 70 |
+
"grad_norm": 8.401183128356934,
|
| 71 |
+
"learning_rate": 6.4142857142857154e-06,
|
| 72 |
+
"loss": 1.378076934814453,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.3843197540353574,
|
| 77 |
+
"grad_norm": 9.6310396194458,
|
| 78 |
+
"learning_rate": 7.1285714285714295e-06,
|
| 79 |
+
"loss": 1.3310549926757813,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.42275172943889316,
|
| 84 |
+
"grad_norm": 9.117944717407227,
|
| 85 |
+
"learning_rate": 7.842857142857143e-06,
|
| 86 |
+
"loss": 1.2295526123046876,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.4611837048424289,
|
| 91 |
+
"grad_norm": 7.5970072746276855,
|
| 92 |
+
"learning_rate": 8.557142857142858e-06,
|
| 93 |
+
"loss": 1.2381634521484375,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.49961568024596464,
|
| 98 |
+
"grad_norm": 8.406278610229492,
|
| 99 |
+
"learning_rate": 9.271428571428573e-06,
|
| 100 |
+
"loss": 1.2207740783691405,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.5380476556495004,
|
| 105 |
+
"grad_norm": 7.469017028808594,
|
| 106 |
+
"learning_rate": 9.985714285714286e-06,
|
| 107 |
+
"loss": 1.2070553588867188,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.5764796310530361,
|
| 112 |
+
"grad_norm": 7.842820167541504,
|
| 113 |
+
"learning_rate": 9.960194963444355e-06,
|
| 114 |
+
"loss": 1.1306779479980469,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.6149116064565718,
|
| 119 |
+
"grad_norm": 8.229959487915039,
|
| 120 |
+
"learning_rate": 9.9195775792039e-06,
|
| 121 |
+
"loss": 1.140580825805664,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.6533435818601077,
|
| 126 |
+
"grad_norm": 7.585798740386963,
|
| 127 |
+
"learning_rate": 9.878960194963446e-06,
|
| 128 |
+
"loss": 1.105696334838867,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.6917755572636434,
|
| 133 |
+
"grad_norm": 7.848736763000488,
|
| 134 |
+
"learning_rate": 9.838342810722991e-06,
|
| 135 |
+
"loss": 1.0822959899902345,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.7302075326671791,
|
| 140 |
+
"grad_norm": 7.353982448577881,
|
| 141 |
+
"learning_rate": 9.797725426482535e-06,
|
| 142 |
+
"loss": 1.0581134796142577,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.7686395080707148,
|
| 147 |
+
"grad_norm": 6.863584518432617,
|
| 148 |
+
"learning_rate": 9.75710804224208e-06,
|
| 149 |
+
"loss": 1.0602525329589845,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.8070714834742506,
|
| 154 |
+
"grad_norm": 7.096811294555664,
|
| 155 |
+
"learning_rate": 9.716490658001624e-06,
|
| 156 |
+
"loss": 1.045466766357422,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.8455034588777863,
|
| 161 |
+
"grad_norm": 7.295511245727539,
|
| 162 |
+
"learning_rate": 9.67587327376117e-06,
|
| 163 |
+
"loss": 1.0288179016113281,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.883935434281322,
|
| 168 |
+
"grad_norm": 6.594845771789551,
|
| 169 |
+
"learning_rate": 9.635255889520715e-06,
|
| 170 |
+
"loss": 1.0182283782958985,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.9223674096848578,
|
| 175 |
+
"grad_norm": 6.083802700042725,
|
| 176 |
+
"learning_rate": 9.594638505280261e-06,
|
| 177 |
+
"loss": 1.0005873107910157,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.9607993850883936,
|
| 182 |
+
"grad_norm": 6.480014801025391,
|
| 183 |
+
"learning_rate": 9.554021121039807e-06,
|
| 184 |
+
"loss": 1.0233656311035155,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.9992313604919293,
|
| 189 |
+
"grad_norm": 7.135806083679199,
|
| 190 |
+
"learning_rate": 9.51340373679935e-06,
|
| 191 |
+
"loss": 1.0027877807617187,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.037663335895465,
|
| 196 |
+
"grad_norm": 7.513897895812988,
|
| 197 |
+
"learning_rate": 9.472786352558896e-06,
|
| 198 |
+
"loss": 0.8653178405761719,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.0760953112990008,
|
| 203 |
+
"grad_norm": 6.412170886993408,
|
| 204 |
+
"learning_rate": 9.432168968318441e-06,
|
| 205 |
+
"loss": 0.8573536682128906,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.1145272867025364,
|
| 210 |
+
"grad_norm": 6.953821659088135,
|
| 211 |
+
"learning_rate": 9.391551584077985e-06,
|
| 212 |
+
"loss": 0.838233871459961,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.1529592621060722,
|
| 217 |
+
"grad_norm": 6.8671650886535645,
|
| 218 |
+
"learning_rate": 9.35093419983753e-06,
|
| 219 |
+
"loss": 0.863470458984375,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.191391237509608,
|
| 224 |
+
"grad_norm": 6.690714359283447,
|
| 225 |
+
"learning_rate": 9.310316815597076e-06,
|
| 226 |
+
"loss": 0.8226670837402343,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.2298232129131437,
|
| 231 |
+
"grad_norm": 5.956531524658203,
|
| 232 |
+
"learning_rate": 9.269699431356622e-06,
|
| 233 |
+
"loss": 0.8588559722900391,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.2682551883166795,
|
| 238 |
+
"grad_norm": 6.430601596832275,
|
| 239 |
+
"learning_rate": 9.229082047116167e-06,
|
| 240 |
+
"loss": 0.8462174224853516,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.3066871637202153,
|
| 245 |
+
"grad_norm": 7.534162998199463,
|
| 246 |
+
"learning_rate": 9.188464662875711e-06,
|
| 247 |
+
"loss": 0.8385597229003906,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.345119139123751,
|
| 252 |
+
"grad_norm": 6.293074131011963,
|
| 253 |
+
"learning_rate": 9.147847278635257e-06,
|
| 254 |
+
"loss": 0.8373613739013672,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.3835511145272867,
|
| 259 |
+
"grad_norm": 6.018751621246338,
|
| 260 |
+
"learning_rate": 9.1072298943948e-06,
|
| 261 |
+
"loss": 0.8304621887207031,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.4219830899308223,
|
| 266 |
+
"grad_norm": 6.324421405792236,
|
| 267 |
+
"learning_rate": 9.066612510154346e-06,
|
| 268 |
+
"loss": 0.8318909454345703,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.4604150653343582,
|
| 273 |
+
"grad_norm": 6.057679653167725,
|
| 274 |
+
"learning_rate": 9.025995125913891e-06,
|
| 275 |
+
"loss": 0.854293212890625,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.498847040737894,
|
| 280 |
+
"grad_norm": 7.707882881164551,
|
| 281 |
+
"learning_rate": 8.985377741673437e-06,
|
| 282 |
+
"loss": 0.8242097473144532,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.5372790161414298,
|
| 287 |
+
"grad_norm": 5.778584957122803,
|
| 288 |
+
"learning_rate": 8.944760357432982e-06,
|
| 289 |
+
"loss": 0.821953125,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.5757109915449654,
|
| 294 |
+
"grad_norm": 5.339688777923584,
|
| 295 |
+
"learning_rate": 8.904142973192526e-06,
|
| 296 |
+
"loss": 0.8104914855957032,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.614142966948501,
|
| 301 |
+
"grad_norm": 5.495669364929199,
|
| 302 |
+
"learning_rate": 8.863525588952072e-06,
|
| 303 |
+
"loss": 0.8245255279541016,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.6525749423520368,
|
| 308 |
+
"grad_norm": 4.540273189544678,
|
| 309 |
+
"learning_rate": 8.822908204711617e-06,
|
| 310 |
+
"loss": 0.819593505859375,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.6910069177555727,
|
| 315 |
+
"grad_norm": 6.755576133728027,
|
| 316 |
+
"learning_rate": 8.782290820471163e-06,
|
| 317 |
+
"loss": 0.8214627838134766,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.7294388931591085,
|
| 322 |
+
"grad_norm": 6.014960289001465,
|
| 323 |
+
"learning_rate": 8.741673436230708e-06,
|
| 324 |
+
"loss": 0.8220201873779297,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.767870868562644,
|
| 329 |
+
"grad_norm": 5.988499641418457,
|
| 330 |
+
"learning_rate": 8.701056051990252e-06,
|
| 331 |
+
"loss": 0.8310813903808594,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.80630284396618,
|
| 336 |
+
"grad_norm": 6.754075050354004,
|
| 337 |
+
"learning_rate": 8.660438667749798e-06,
|
| 338 |
+
"loss": 0.8343869781494141,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.8447348193697155,
|
| 343 |
+
"grad_norm": 5.107484817504883,
|
| 344 |
+
"learning_rate": 8.619821283509343e-06,
|
| 345 |
+
"loss": 0.8037309265136718,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.8831667947732513,
|
| 350 |
+
"grad_norm": 6.787992477416992,
|
| 351 |
+
"learning_rate": 8.579203899268887e-06,
|
| 352 |
+
"loss": 0.8196144866943359,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.9215987701767872,
|
| 357 |
+
"grad_norm": 5.575016021728516,
|
| 358 |
+
"learning_rate": 8.538586515028433e-06,
|
| 359 |
+
"loss": 0.7984078216552735,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.960030745580323,
|
| 364 |
+
"grad_norm": 5.993164539337158,
|
| 365 |
+
"learning_rate": 8.497969130787978e-06,
|
| 366 |
+
"loss": 0.8131062316894532,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.9984627209838586,
|
| 371 |
+
"grad_norm": 5.438976287841797,
|
| 372 |
+
"learning_rate": 8.457351746547524e-06,
|
| 373 |
+
"loss": 0.8112478637695313,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.036894696387394,
|
| 378 |
+
"grad_norm": 5.135393142700195,
|
| 379 |
+
"learning_rate": 8.416734362307069e-06,
|
| 380 |
+
"loss": 0.6605772399902343,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.07532667179093,
|
| 385 |
+
"grad_norm": 5.112312316894531,
|
| 386 |
+
"learning_rate": 8.376116978066613e-06,
|
| 387 |
+
"loss": 0.6184328079223633,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 2.113758647194466,
|
| 392 |
+
"grad_norm": 4.906278610229492,
|
| 393 |
+
"learning_rate": 8.335499593826158e-06,
|
| 394 |
+
"loss": 0.6375198745727539,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.1521906225980016,
|
| 399 |
+
"grad_norm": 6.010103702545166,
|
| 400 |
+
"learning_rate": 8.294882209585702e-06,
|
| 401 |
+
"loss": 0.6402740478515625,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 2.1906225980015375,
|
| 406 |
+
"grad_norm": 5.4578423500061035,
|
| 407 |
+
"learning_rate": 8.254264825345248e-06,
|
| 408 |
+
"loss": 0.6382756042480469,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.229054573405073,
|
| 413 |
+
"grad_norm": 4.880190849304199,
|
| 414 |
+
"learning_rate": 8.213647441104793e-06,
|
| 415 |
+
"loss": 0.6498109436035157,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 2.2674865488086087,
|
| 420 |
+
"grad_norm": 4.855320453643799,
|
| 421 |
+
"learning_rate": 8.173030056864339e-06,
|
| 422 |
+
"loss": 0.6391303253173828,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.3059185242121445,
|
| 427 |
+
"grad_norm": 5.620919704437256,
|
| 428 |
+
"learning_rate": 8.132412672623884e-06,
|
| 429 |
+
"loss": 0.642280502319336,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.3443504996156803,
|
| 434 |
+
"grad_norm": 4.731573581695557,
|
| 435 |
+
"learning_rate": 8.091795288383428e-06,
|
| 436 |
+
"loss": 0.6625186157226562,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 2.382782475019216,
|
| 441 |
+
"grad_norm": 5.539109706878662,
|
| 442 |
+
"learning_rate": 8.051177904142974e-06,
|
| 443 |
+
"loss": 0.6410211181640625,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.4212144504227515,
|
| 448 |
+
"grad_norm": 4.798823356628418,
|
| 449 |
+
"learning_rate": 8.01056051990252e-06,
|
| 450 |
+
"loss": 0.6538510894775391,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.4596464258262873,
|
| 455 |
+
"grad_norm": 5.686472415924072,
|
| 456 |
+
"learning_rate": 7.969943135662063e-06,
|
| 457 |
+
"loss": 0.6487322235107422,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 2.498078401229823,
|
| 462 |
+
"grad_norm": 4.942172050476074,
|
| 463 |
+
"learning_rate": 7.929325751421608e-06,
|
| 464 |
+
"loss": 0.6391149139404297,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.536510376633359,
|
| 469 |
+
"grad_norm": 5.06510591506958,
|
| 470 |
+
"learning_rate": 7.888708367181154e-06,
|
| 471 |
+
"loss": 0.6553841400146484,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 2.574942352036895,
|
| 476 |
+
"grad_norm": 5.525066375732422,
|
| 477 |
+
"learning_rate": 7.8480909829407e-06,
|
| 478 |
+
"loss": 0.6518791198730469,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 2.6133743274404306,
|
| 483 |
+
"grad_norm": 4.838707447052002,
|
| 484 |
+
"learning_rate": 7.807473598700245e-06,
|
| 485 |
+
"loss": 0.6638088989257812,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 2.6518063028439665,
|
| 490 |
+
"grad_norm": 5.3868255615234375,
|
| 491 |
+
"learning_rate": 7.766856214459789e-06,
|
| 492 |
+
"loss": 0.6488023376464844,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 2.690238278247502,
|
| 497 |
+
"grad_norm": 5.59976863861084,
|
| 498 |
+
"learning_rate": 7.726238830219334e-06,
|
| 499 |
+
"loss": 0.6557725524902344,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 2.7286702536510377,
|
| 504 |
+
"grad_norm": 5.351370811462402,
|
| 505 |
+
"learning_rate": 7.68562144597888e-06,
|
| 506 |
+
"loss": 0.647769546508789,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 2.7671022290545735,
|
| 511 |
+
"grad_norm": 4.642946243286133,
|
| 512 |
+
"learning_rate": 7.645004061738425e-06,
|
| 513 |
+
"loss": 0.6705641174316406,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 2.8055342044581093,
|
| 518 |
+
"grad_norm": 5.362582683563232,
|
| 519 |
+
"learning_rate": 7.60438667749797e-06,
|
| 520 |
+
"loss": 0.6718820190429687,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 2.8439661798616447,
|
| 525 |
+
"grad_norm": 5.851658344268799,
|
| 526 |
+
"learning_rate": 7.563769293257515e-06,
|
| 527 |
+
"loss": 0.637021713256836,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.8823981552651805,
|
| 532 |
+
"grad_norm": 4.618478298187256,
|
| 533 |
+
"learning_rate": 7.52315190901706e-06,
|
| 534 |
+
"loss": 0.638284797668457,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.9208301306687163,
|
| 539 |
+
"grad_norm": 5.931302547454834,
|
| 540 |
+
"learning_rate": 7.482534524776605e-06,
|
| 541 |
+
"loss": 0.6481784057617187,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.959262106072252,
|
| 546 |
+
"grad_norm": 4.9440717697143555,
|
| 547 |
+
"learning_rate": 7.4419171405361504e-06,
|
| 548 |
+
"loss": 0.6746251678466797,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.997694081475788,
|
| 553 |
+
"grad_norm": 5.484557151794434,
|
| 554 |
+
"learning_rate": 7.401299756295696e-06,
|
| 555 |
+
"loss": 0.6457685089111328,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 3.036126056879324,
|
| 560 |
+
"grad_norm": 4.5576581954956055,
|
| 561 |
+
"learning_rate": 7.36068237205524e-06,
|
| 562 |
+
"loss": 0.5101409149169922,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 3.074558032282859,
|
| 567 |
+
"grad_norm": 4.8531646728515625,
|
| 568 |
+
"learning_rate": 7.320064987814785e-06,
|
| 569 |
+
"loss": 0.5025056457519531,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 3.112990007686395,
|
| 574 |
+
"grad_norm": 4.5276618003845215,
|
| 575 |
+
"learning_rate": 7.27944760357433e-06,
|
| 576 |
+
"loss": 0.49780776977539065,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 3.151421983089931,
|
| 581 |
+
"grad_norm": 5.236543655395508,
|
| 582 |
+
"learning_rate": 7.2388302193338755e-06,
|
| 583 |
+
"loss": 0.49986991882324217,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 3.1898539584934666,
|
| 588 |
+
"grad_norm": 5.309688568115234,
|
| 589 |
+
"learning_rate": 7.198212835093421e-06,
|
| 590 |
+
"loss": 0.5006533813476562,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 3.2282859338970025,
|
| 595 |
+
"grad_norm": 4.870418071746826,
|
| 596 |
+
"learning_rate": 7.157595450852966e-06,
|
| 597 |
+
"loss": 0.49584457397460935,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 3.266717909300538,
|
| 602 |
+
"grad_norm": 4.612658500671387,
|
| 603 |
+
"learning_rate": 7.116978066612511e-06,
|
| 604 |
+
"loss": 0.5062141418457031,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 3.3051498847040737,
|
| 609 |
+
"grad_norm": 5.27450704574585,
|
| 610 |
+
"learning_rate": 7.076360682372055e-06,
|
| 611 |
+
"loss": 0.4989910507202148,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 3.3435818601076095,
|
| 616 |
+
"grad_norm": 4.7967023849487305,
|
| 617 |
+
"learning_rate": 7.035743298131601e-06,
|
| 618 |
+
"loss": 0.5004584503173828,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 3.3820138355111453,
|
| 623 |
+
"grad_norm": 4.755988597869873,
|
| 624 |
+
"learning_rate": 6.995125913891147e-06,
|
| 625 |
+
"loss": 0.5030863952636718,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 3.420445810914681,
|
| 630 |
+
"grad_norm": 4.493087291717529,
|
| 631 |
+
"learning_rate": 6.954508529650691e-06,
|
| 632 |
+
"loss": 0.5036092376708985,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.458877786318217,
|
| 637 |
+
"grad_norm": 4.850837707519531,
|
| 638 |
+
"learning_rate": 6.913891145410236e-06,
|
| 639 |
+
"loss": 0.509452896118164,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 3.4973097617217523,
|
| 644 |
+
"grad_norm": 4.727443218231201,
|
| 645 |
+
"learning_rate": 6.873273761169781e-06,
|
| 646 |
+
"loss": 0.5102983474731445,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 3.535741737125288,
|
| 651 |
+
"grad_norm": 4.911144256591797,
|
| 652 |
+
"learning_rate": 6.832656376929326e-06,
|
| 653 |
+
"loss": 0.5117542266845703,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.574173712528824,
|
| 658 |
+
"grad_norm": 4.4287872314453125,
|
| 659 |
+
"learning_rate": 6.792038992688872e-06,
|
| 660 |
+
"loss": 0.5030385208129883,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 3.61260568793236,
|
| 665 |
+
"grad_norm": 5.549263954162598,
|
| 666 |
+
"learning_rate": 6.751421608448417e-06,
|
| 667 |
+
"loss": 0.5239413833618164,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.651037663335895,
|
| 672 |
+
"grad_norm": 4.574188709259033,
|
| 673 |
+
"learning_rate": 6.710804224207962e-06,
|
| 674 |
+
"loss": 0.5226581573486329,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 3.689469638739431,
|
| 679 |
+
"grad_norm": 5.468628406524658,
|
| 680 |
+
"learning_rate": 6.670186839967506e-06,
|
| 681 |
+
"loss": 0.5152123260498047,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 3.727901614142967,
|
| 686 |
+
"grad_norm": 5.479881763458252,
|
| 687 |
+
"learning_rate": 6.6295694557270514e-06,
|
| 688 |
+
"loss": 0.5082434844970704,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 3.7663335895465027,
|
| 693 |
+
"grad_norm": 4.840885162353516,
|
| 694 |
+
"learning_rate": 6.588952071486597e-06,
|
| 695 |
+
"loss": 0.5003246307373047,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 3.8047655649500385,
|
| 700 |
+
"grad_norm": 4.748307704925537,
|
| 701 |
+
"learning_rate": 6.548334687246142e-06,
|
| 702 |
+
"loss": 0.5111896514892578,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 3.8431975403535743,
|
| 707 |
+
"grad_norm": 5.050412178039551,
|
| 708 |
+
"learning_rate": 6.507717303005687e-06,
|
| 709 |
+
"loss": 0.5184495162963867,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 3.88162951575711,
|
| 714 |
+
"grad_norm": 5.65321159362793,
|
| 715 |
+
"learning_rate": 6.467099918765232e-06,
|
| 716 |
+
"loss": 0.5129562377929687,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.9200614911606455,
|
| 721 |
+
"grad_norm": 4.804498195648193,
|
| 722 |
+
"learning_rate": 6.426482534524777e-06,
|
| 723 |
+
"loss": 0.5197629928588867,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 3.9584934665641813,
|
| 728 |
+
"grad_norm": 5.435920715332031,
|
| 729 |
+
"learning_rate": 6.385865150284323e-06,
|
| 730 |
+
"loss": 0.5145494079589844,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 3.996925441967717,
|
| 735 |
+
"grad_norm": 4.372884750366211,
|
| 736 |
+
"learning_rate": 6.3452477660438675e-06,
|
| 737 |
+
"loss": 0.5092922973632813,
|
| 738 |
+
"step": 5200
|
| 739 |
+
}
|
| 740 |
+
],
|
| 741 |
+
"logging_steps": 50,
|
| 742 |
+
"max_steps": 13010,
|
| 743 |
+
"num_input_tokens_seen": 0,
|
| 744 |
+
"num_train_epochs": 10,
|
| 745 |
+
"save_steps": 500,
|
| 746 |
+
"stateful_callbacks": {
|
| 747 |
+
"TrainerControl": {
|
| 748 |
+
"args": {
|
| 749 |
+
"should_epoch_stop": false,
|
| 750 |
+
"should_evaluate": false,
|
| 751 |
+
"should_log": false,
|
| 752 |
+
"should_save": true,
|
| 753 |
+
"should_training_stop": false
|
| 754 |
+
},
|
| 755 |
+
"attributes": {}
|
| 756 |
+
}
|
| 757 |
+
},
|
| 758 |
+
"total_flos": 6.798371835251733e+20,
|
| 759 |
+
"train_batch_size": 16,
|
| 760 |
+
"trial_name": null,
|
| 761 |
+
"trial_params": null
|
| 762 |
+
}
|
checkpoint-5204/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ee3bcd14a0b55278a89671dd7c00d9c4f80d655c6b999121882d6d90242f8b0
|
| 3 |
+
size 5329
|