Upload 8 files
Browse files- config.json +48 -0
- generation_config.json +257 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
- preprocessor_config.json +15 -0
- trainer_state.json +1077 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
50256
|
| 12 |
+
],
|
| 13 |
+
"bos_token_id": 50257,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"d_model": 1280,
|
| 16 |
+
"decoder_attention_heads": 20,
|
| 17 |
+
"decoder_ffn_dim": 5120,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 32,
|
| 20 |
+
"decoder_start_token_id": 50258,
|
| 21 |
+
"dropout": 0.0,
|
| 22 |
+
"dtype": "float32",
|
| 23 |
+
"encoder_attention_heads": 20,
|
| 24 |
+
"encoder_ffn_dim": 5120,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 32,
|
| 27 |
+
"eos_token_id": 50257,
|
| 28 |
+
"init_std": 0.02,
|
| 29 |
+
"is_encoder_decoder": true,
|
| 30 |
+
"mask_feature_length": 10,
|
| 31 |
+
"mask_feature_min_masks": 0,
|
| 32 |
+
"mask_feature_prob": 0.0,
|
| 33 |
+
"mask_time_length": 10,
|
| 34 |
+
"mask_time_min_masks": 2,
|
| 35 |
+
"mask_time_prob": 0.05,
|
| 36 |
+
"max_source_positions": 1500,
|
| 37 |
+
"max_target_positions": 448,
|
| 38 |
+
"median_filter_width": 7,
|
| 39 |
+
"model_type": "whisper",
|
| 40 |
+
"num_hidden_layers": 32,
|
| 41 |
+
"num_mel_bins": 128,
|
| 42 |
+
"pad_token_id": 50256,
|
| 43 |
+
"scale_embedding": false,
|
| 44 |
+
"transformers_version": "4.57.3",
|
| 45 |
+
"use_cache": false,
|
| 46 |
+
"use_weighted_layer_sum": false,
|
| 47 |
+
"vocab_size": 51866
|
| 48 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,257 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alignment_heads": [
|
| 3 |
+
[
|
| 4 |
+
7,
|
| 5 |
+
0
|
| 6 |
+
],
|
| 7 |
+
[
|
| 8 |
+
10,
|
| 9 |
+
17
|
| 10 |
+
],
|
| 11 |
+
[
|
| 12 |
+
12,
|
| 13 |
+
18
|
| 14 |
+
],
|
| 15 |
+
[
|
| 16 |
+
13,
|
| 17 |
+
12
|
| 18 |
+
],
|
| 19 |
+
[
|
| 20 |
+
16,
|
| 21 |
+
1
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
17,
|
| 25 |
+
14
|
| 26 |
+
],
|
| 27 |
+
[
|
| 28 |
+
19,
|
| 29 |
+
11
|
| 30 |
+
],
|
| 31 |
+
[
|
| 32 |
+
21,
|
| 33 |
+
4
|
| 34 |
+
],
|
| 35 |
+
[
|
| 36 |
+
24,
|
| 37 |
+
1
|
| 38 |
+
],
|
| 39 |
+
[
|
| 40 |
+
25,
|
| 41 |
+
6
|
| 42 |
+
]
|
| 43 |
+
],
|
| 44 |
+
"begin_suppress_tokens": [
|
| 45 |
+
220
|
| 46 |
+
],
|
| 47 |
+
"bos_token_id": 50257,
|
| 48 |
+
"decoder_start_token_id": 50258,
|
| 49 |
+
"eos_token_id": 50257,
|
| 50 |
+
"forced_decoder_ids": null,
|
| 51 |
+
"is_multilingual": true,
|
| 52 |
+
"lang_to_id": {
|
| 53 |
+
"<|af|>": 50327,
|
| 54 |
+
"<|am|>": 50334,
|
| 55 |
+
"<|ar|>": 50272,
|
| 56 |
+
"<|as|>": 50350,
|
| 57 |
+
"<|az|>": 50304,
|
| 58 |
+
"<|ba|>": 50355,
|
| 59 |
+
"<|be|>": 50330,
|
| 60 |
+
"<|bg|>": 50292,
|
| 61 |
+
"<|bn|>": 50302,
|
| 62 |
+
"<|bo|>": 50347,
|
| 63 |
+
"<|br|>": 50309,
|
| 64 |
+
"<|bs|>": 50315,
|
| 65 |
+
"<|ca|>": 50270,
|
| 66 |
+
"<|cs|>": 50283,
|
| 67 |
+
"<|cy|>": 50297,
|
| 68 |
+
"<|da|>": 50285,
|
| 69 |
+
"<|de|>": 50261,
|
| 70 |
+
"<|el|>": 50281,
|
| 71 |
+
"<|en|>": 50259,
|
| 72 |
+
"<|es|>": 50262,
|
| 73 |
+
"<|et|>": 50307,
|
| 74 |
+
"<|eu|>": 50310,
|
| 75 |
+
"<|fa|>": 50300,
|
| 76 |
+
"<|fi|>": 50277,
|
| 77 |
+
"<|fo|>": 50338,
|
| 78 |
+
"<|fr|>": 50265,
|
| 79 |
+
"<|gl|>": 50319,
|
| 80 |
+
"<|gu|>": 50333,
|
| 81 |
+
"<|haw|>": 50352,
|
| 82 |
+
"<|ha|>": 50354,
|
| 83 |
+
"<|he|>": 50279,
|
| 84 |
+
"<|hi|>": 50276,
|
| 85 |
+
"<|hr|>": 50291,
|
| 86 |
+
"<|ht|>": 50339,
|
| 87 |
+
"<|hu|>": 50286,
|
| 88 |
+
"<|hy|>": 50312,
|
| 89 |
+
"<|id|>": 50275,
|
| 90 |
+
"<|is|>": 50311,
|
| 91 |
+
"<|it|>": 50274,
|
| 92 |
+
"<|ja|>": 50266,
|
| 93 |
+
"<|jw|>": 50356,
|
| 94 |
+
"<|ka|>": 50329,
|
| 95 |
+
"<|kk|>": 50316,
|
| 96 |
+
"<|km|>": 50323,
|
| 97 |
+
"<|kn|>": 50306,
|
| 98 |
+
"<|ko|>": 50264,
|
| 99 |
+
"<|la|>": 50294,
|
| 100 |
+
"<|lb|>": 50345,
|
| 101 |
+
"<|ln|>": 50353,
|
| 102 |
+
"<|lo|>": 50336,
|
| 103 |
+
"<|lt|>": 50293,
|
| 104 |
+
"<|lv|>": 50301,
|
| 105 |
+
"<|mg|>": 50349,
|
| 106 |
+
"<|mi|>": 50295,
|
| 107 |
+
"<|mk|>": 50308,
|
| 108 |
+
"<|ml|>": 50296,
|
| 109 |
+
"<|mn|>": 50314,
|
| 110 |
+
"<|mr|>": 50320,
|
| 111 |
+
"<|ms|>": 50282,
|
| 112 |
+
"<|mt|>": 50343,
|
| 113 |
+
"<|my|>": 50346,
|
| 114 |
+
"<|ne|>": 50313,
|
| 115 |
+
"<|nl|>": 50271,
|
| 116 |
+
"<|nn|>": 50342,
|
| 117 |
+
"<|no|>": 50288,
|
| 118 |
+
"<|oc|>": 50328,
|
| 119 |
+
"<|pa|>": 50321,
|
| 120 |
+
"<|pl|>": 50269,
|
| 121 |
+
"<|ps|>": 50340,
|
| 122 |
+
"<|pt|>": 50267,
|
| 123 |
+
"<|ro|>": 50284,
|
| 124 |
+
"<|ru|>": 50263,
|
| 125 |
+
"<|sa|>": 50344,
|
| 126 |
+
"<|sd|>": 50332,
|
| 127 |
+
"<|si|>": 50322,
|
| 128 |
+
"<|sk|>": 50298,
|
| 129 |
+
"<|sl|>": 50305,
|
| 130 |
+
"<|sn|>": 50324,
|
| 131 |
+
"<|so|>": 50326,
|
| 132 |
+
"<|sq|>": 50317,
|
| 133 |
+
"<|sr|>": 50303,
|
| 134 |
+
"<|su|>": 50357,
|
| 135 |
+
"<|sv|>": 50273,
|
| 136 |
+
"<|sw|>": 50318,
|
| 137 |
+
"<|ta|>": 50287,
|
| 138 |
+
"<|te|>": 50299,
|
| 139 |
+
"<|tg|>": 50331,
|
| 140 |
+
"<|th|>": 50289,
|
| 141 |
+
"<|tk|>": 50341,
|
| 142 |
+
"<|tl|>": 50348,
|
| 143 |
+
"<|tr|>": 50268,
|
| 144 |
+
"<|tt|>": 50351,
|
| 145 |
+
"<|uk|>": 50280,
|
| 146 |
+
"<|ur|>": 50290,
|
| 147 |
+
"<|uz|>": 50337,
|
| 148 |
+
"<|vi|>": 50278,
|
| 149 |
+
"<|yi|>": 50335,
|
| 150 |
+
"<|yo|>": 50325,
|
| 151 |
+
"<|yue|>": 50358,
|
| 152 |
+
"<|zh|>": 50260
|
| 153 |
+
},
|
| 154 |
+
"language": "english",
|
| 155 |
+
"max_initial_timestamp_index": 50,
|
| 156 |
+
"max_length": 448,
|
| 157 |
+
"no_timestamps_token_id": 50364,
|
| 158 |
+
"pad_token_id": 50257,
|
| 159 |
+
"prev_sot_token_id": 50362,
|
| 160 |
+
"return_timestamps": false,
|
| 161 |
+
"suppress_tokens": [
|
| 162 |
+
1,
|
| 163 |
+
2,
|
| 164 |
+
7,
|
| 165 |
+
8,
|
| 166 |
+
9,
|
| 167 |
+
10,
|
| 168 |
+
14,
|
| 169 |
+
25,
|
| 170 |
+
26,
|
| 171 |
+
27,
|
| 172 |
+
28,
|
| 173 |
+
29,
|
| 174 |
+
31,
|
| 175 |
+
58,
|
| 176 |
+
59,
|
| 177 |
+
60,
|
| 178 |
+
61,
|
| 179 |
+
62,
|
| 180 |
+
63,
|
| 181 |
+
90,
|
| 182 |
+
91,
|
| 183 |
+
92,
|
| 184 |
+
93,
|
| 185 |
+
359,
|
| 186 |
+
503,
|
| 187 |
+
522,
|
| 188 |
+
542,
|
| 189 |
+
873,
|
| 190 |
+
893,
|
| 191 |
+
902,
|
| 192 |
+
918,
|
| 193 |
+
922,
|
| 194 |
+
931,
|
| 195 |
+
1350,
|
| 196 |
+
1853,
|
| 197 |
+
1982,
|
| 198 |
+
2460,
|
| 199 |
+
2627,
|
| 200 |
+
3246,
|
| 201 |
+
3253,
|
| 202 |
+
3268,
|
| 203 |
+
3536,
|
| 204 |
+
3846,
|
| 205 |
+
3961,
|
| 206 |
+
4183,
|
| 207 |
+
4667,
|
| 208 |
+
6585,
|
| 209 |
+
6647,
|
| 210 |
+
7273,
|
| 211 |
+
9061,
|
| 212 |
+
9383,
|
| 213 |
+
10428,
|
| 214 |
+
10929,
|
| 215 |
+
11938,
|
| 216 |
+
12033,
|
| 217 |
+
12331,
|
| 218 |
+
12562,
|
| 219 |
+
13793,
|
| 220 |
+
14157,
|
| 221 |
+
14635,
|
| 222 |
+
15265,
|
| 223 |
+
15618,
|
| 224 |
+
16553,
|
| 225 |
+
16604,
|
| 226 |
+
18362,
|
| 227 |
+
18956,
|
| 228 |
+
20075,
|
| 229 |
+
21675,
|
| 230 |
+
22520,
|
| 231 |
+
26130,
|
| 232 |
+
26161,
|
| 233 |
+
26435,
|
| 234 |
+
28279,
|
| 235 |
+
29464,
|
| 236 |
+
31650,
|
| 237 |
+
32302,
|
| 238 |
+
32470,
|
| 239 |
+
36865,
|
| 240 |
+
42863,
|
| 241 |
+
47425,
|
| 242 |
+
49870,
|
| 243 |
+
50254,
|
| 244 |
+
50258,
|
| 245 |
+
50359,
|
| 246 |
+
50360,
|
| 247 |
+
50361,
|
| 248 |
+
50362,
|
| 249 |
+
50363
|
| 250 |
+
],
|
| 251 |
+
"task": "transcribe",
|
| 252 |
+
"task_to_id": {
|
| 253 |
+
"transcribe": 50360,
|
| 254 |
+
"translate": 50359
|
| 255 |
+
},
|
| 256 |
+
"transformers_version": "4.57.3"
|
| 257 |
+
}
|
model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3692843de11f9d606757aaa8ecd478394f2518693ec4e2d392eaef8ca4d2e318
|
| 3 |
+
size 4993448880
|
model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c4376eb52a19a3527ceb2d7ca30db1fe32510ad8e16cc4e8f751c54b5db945e0
|
| 3 |
+
size 1180663192
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"dither": 0.0,
|
| 4 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 5 |
+
"feature_size": 128,
|
| 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 |
+
"processor_class": "WhisperProcessor",
|
| 13 |
+
"return_attention_mask": false,
|
| 14 |
+
"sampling_rate": 16000
|
| 15 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1077 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3500,
|
| 3 |
+
"best_metric": 14.107423500046224,
|
| 4 |
+
"best_model_checkpoint": "./whisper-large-v3-atc-mrezzat/checkpoint-3500",
|
| 5 |
+
"epoch": 7.526881720430108,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.053763440860215055,
|
| 14 |
+
"grad_norm": 7.112494945526123,
|
| 15 |
+
"learning_rate": 4.800000000000001e-07,
|
| 16 |
+
"loss": 1.2801,
|
| 17 |
+
"step": 25
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.10752688172043011,
|
| 21 |
+
"grad_norm": 3.566549062728882,
|
| 22 |
+
"learning_rate": 9.800000000000001e-07,
|
| 23 |
+
"loss": 1.0244,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.16129032258064516,
|
| 28 |
+
"grad_norm": 3.967959403991699,
|
| 29 |
+
"learning_rate": 1.48e-06,
|
| 30 |
+
"loss": 0.8328,
|
| 31 |
+
"step": 75
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.21505376344086022,
|
| 35 |
+
"grad_norm": 4.964681625366211,
|
| 36 |
+
"learning_rate": 1.98e-06,
|
| 37 |
+
"loss": 0.7712,
|
| 38 |
+
"step": 100
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.26881720430107525,
|
| 42 |
+
"grad_norm": 3.7016031742095947,
|
| 43 |
+
"learning_rate": 2.4800000000000004e-06,
|
| 44 |
+
"loss": 0.684,
|
| 45 |
+
"step": 125
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.3225806451612903,
|
| 49 |
+
"grad_norm": 2.4758129119873047,
|
| 50 |
+
"learning_rate": 2.9800000000000003e-06,
|
| 51 |
+
"loss": 0.6756,
|
| 52 |
+
"step": 150
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.3763440860215054,
|
| 56 |
+
"grad_norm": 4.086286544799805,
|
| 57 |
+
"learning_rate": 3.48e-06,
|
| 58 |
+
"loss": 0.6284,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.43010752688172044,
|
| 63 |
+
"grad_norm": 3.4064314365386963,
|
| 64 |
+
"learning_rate": 3.980000000000001e-06,
|
| 65 |
+
"loss": 0.5733,
|
| 66 |
+
"step": 200
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.4838709677419355,
|
| 70 |
+
"grad_norm": 3.6095917224884033,
|
| 71 |
+
"learning_rate": 4.48e-06,
|
| 72 |
+
"loss": 0.5348,
|
| 73 |
+
"step": 225
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.5376344086021505,
|
| 77 |
+
"grad_norm": 3.179569959640503,
|
| 78 |
+
"learning_rate": 4.980000000000001e-06,
|
| 79 |
+
"loss": 0.5174,
|
| 80 |
+
"step": 250
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.5913978494623656,
|
| 84 |
+
"grad_norm": 2.937546491622925,
|
| 85 |
+
"learning_rate": 5.480000000000001e-06,
|
| 86 |
+
"loss": 0.4462,
|
| 87 |
+
"step": 275
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.6451612903225806,
|
| 91 |
+
"grad_norm": 3.4652342796325684,
|
| 92 |
+
"learning_rate": 5.98e-06,
|
| 93 |
+
"loss": 0.5151,
|
| 94 |
+
"step": 300
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.6989247311827957,
|
| 98 |
+
"grad_norm": 3.50107479095459,
|
| 99 |
+
"learning_rate": 6.480000000000001e-06,
|
| 100 |
+
"loss": 0.466,
|
| 101 |
+
"step": 325
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.7526881720430108,
|
| 105 |
+
"grad_norm": 2.7846341133117676,
|
| 106 |
+
"learning_rate": 6.98e-06,
|
| 107 |
+
"loss": 0.4724,
|
| 108 |
+
"step": 350
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.8064516129032258,
|
| 112 |
+
"grad_norm": 2.7892351150512695,
|
| 113 |
+
"learning_rate": 7.48e-06,
|
| 114 |
+
"loss": 0.4458,
|
| 115 |
+
"step": 375
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.8602150537634409,
|
| 119 |
+
"grad_norm": 3.1615209579467773,
|
| 120 |
+
"learning_rate": 7.980000000000002e-06,
|
| 121 |
+
"loss": 0.508,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.9139784946236559,
|
| 126 |
+
"grad_norm": 2.775057792663574,
|
| 127 |
+
"learning_rate": 8.48e-06,
|
| 128 |
+
"loss": 0.4454,
|
| 129 |
+
"step": 425
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.967741935483871,
|
| 133 |
+
"grad_norm": 3.7355740070343018,
|
| 134 |
+
"learning_rate": 8.98e-06,
|
| 135 |
+
"loss": 0.4455,
|
| 136 |
+
"step": 450
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.021505376344086,
|
| 140 |
+
"grad_norm": 3.1661925315856934,
|
| 141 |
+
"learning_rate": 9.48e-06,
|
| 142 |
+
"loss": 0.3986,
|
| 143 |
+
"step": 475
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.075268817204301,
|
| 147 |
+
"grad_norm": 2.5334320068359375,
|
| 148 |
+
"learning_rate": 9.980000000000001e-06,
|
| 149 |
+
"loss": 0.3733,
|
| 150 |
+
"step": 500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.075268817204301,
|
| 154 |
+
"eval_loss": 0.2843235731124878,
|
| 155 |
+
"eval_runtime": 649.0922,
|
| 156 |
+
"eval_samples_per_second": 1.457,
|
| 157 |
+
"eval_steps_per_second": 1.457,
|
| 158 |
+
"eval_wer": 16.54802625496903,
|
| 159 |
+
"step": 500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.129032258064516,
|
| 163 |
+
"grad_norm": 1.8294272422790527,
|
| 164 |
+
"learning_rate": 9.942168674698795e-06,
|
| 165 |
+
"loss": 0.3758,
|
| 166 |
+
"step": 525
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.1827956989247312,
|
| 170 |
+
"grad_norm": 2.663872718811035,
|
| 171 |
+
"learning_rate": 9.881927710843375e-06,
|
| 172 |
+
"loss": 0.3543,
|
| 173 |
+
"step": 550
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.2365591397849462,
|
| 177 |
+
"grad_norm": 3.4595935344696045,
|
| 178 |
+
"learning_rate": 9.821686746987952e-06,
|
| 179 |
+
"loss": 0.3337,
|
| 180 |
+
"step": 575
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.2903225806451613,
|
| 184 |
+
"grad_norm": 3.2685301303863525,
|
| 185 |
+
"learning_rate": 9.76144578313253e-06,
|
| 186 |
+
"loss": 0.3369,
|
| 187 |
+
"step": 600
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.3440860215053765,
|
| 191 |
+
"grad_norm": 3.0888893604278564,
|
| 192 |
+
"learning_rate": 9.70120481927711e-06,
|
| 193 |
+
"loss": 0.3624,
|
| 194 |
+
"step": 625
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.3978494623655915,
|
| 198 |
+
"grad_norm": 2.6340599060058594,
|
| 199 |
+
"learning_rate": 9.640963855421687e-06,
|
| 200 |
+
"loss": 0.3562,
|
| 201 |
+
"step": 650
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.4516129032258065,
|
| 205 |
+
"grad_norm": 1.5518828630447388,
|
| 206 |
+
"learning_rate": 9.580722891566265e-06,
|
| 207 |
+
"loss": 0.3335,
|
| 208 |
+
"step": 675
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.5053763440860215,
|
| 212 |
+
"grad_norm": 2.7814254760742188,
|
| 213 |
+
"learning_rate": 9.520481927710844e-06,
|
| 214 |
+
"loss": 0.3385,
|
| 215 |
+
"step": 700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.5591397849462365,
|
| 219 |
+
"grad_norm": 2.7169201374053955,
|
| 220 |
+
"learning_rate": 9.460240963855422e-06,
|
| 221 |
+
"loss": 0.3304,
|
| 222 |
+
"step": 725
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.6129032258064515,
|
| 226 |
+
"grad_norm": 2.4786252975463867,
|
| 227 |
+
"learning_rate": 9.4e-06,
|
| 228 |
+
"loss": 0.3158,
|
| 229 |
+
"step": 750
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.6666666666666665,
|
| 233 |
+
"grad_norm": 2.5731842517852783,
|
| 234 |
+
"learning_rate": 9.33975903614458e-06,
|
| 235 |
+
"loss": 0.3306,
|
| 236 |
+
"step": 775
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 1.7204301075268817,
|
| 240 |
+
"grad_norm": 3.4572856426239014,
|
| 241 |
+
"learning_rate": 9.279518072289157e-06,
|
| 242 |
+
"loss": 0.3023,
|
| 243 |
+
"step": 800
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.7741935483870968,
|
| 247 |
+
"grad_norm": 2.8618459701538086,
|
| 248 |
+
"learning_rate": 9.219277108433736e-06,
|
| 249 |
+
"loss": 0.325,
|
| 250 |
+
"step": 825
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.827956989247312,
|
| 254 |
+
"grad_norm": 3.5112569332122803,
|
| 255 |
+
"learning_rate": 9.159036144578314e-06,
|
| 256 |
+
"loss": 0.3206,
|
| 257 |
+
"step": 850
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.881720430107527,
|
| 261 |
+
"grad_norm": 2.5521082878112793,
|
| 262 |
+
"learning_rate": 9.098795180722892e-06,
|
| 263 |
+
"loss": 0.285,
|
| 264 |
+
"step": 875
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.935483870967742,
|
| 268 |
+
"grad_norm": 3.105656623840332,
|
| 269 |
+
"learning_rate": 9.038554216867471e-06,
|
| 270 |
+
"loss": 0.2805,
|
| 271 |
+
"step": 900
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.989247311827957,
|
| 275 |
+
"grad_norm": 3.3313686847686768,
|
| 276 |
+
"learning_rate": 8.978313253012049e-06,
|
| 277 |
+
"loss": 0.3012,
|
| 278 |
+
"step": 925
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.043010752688172,
|
| 282 |
+
"grad_norm": 2.5710561275482178,
|
| 283 |
+
"learning_rate": 8.918072289156628e-06,
|
| 284 |
+
"loss": 0.2348,
|
| 285 |
+
"step": 950
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.096774193548387,
|
| 289 |
+
"grad_norm": 2.4275355339050293,
|
| 290 |
+
"learning_rate": 8.857831325301206e-06,
|
| 291 |
+
"loss": 0.1777,
|
| 292 |
+
"step": 975
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.150537634408602,
|
| 296 |
+
"grad_norm": 1.8588894605636597,
|
| 297 |
+
"learning_rate": 8.797590361445784e-06,
|
| 298 |
+
"loss": 0.1973,
|
| 299 |
+
"step": 1000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.150537634408602,
|
| 303 |
+
"eval_loss": 0.28681960701942444,
|
| 304 |
+
"eval_runtime": 657.9873,
|
| 305 |
+
"eval_samples_per_second": 1.438,
|
| 306 |
+
"eval_steps_per_second": 1.438,
|
| 307 |
+
"eval_wer": 14.588148285106776,
|
| 308 |
+
"step": 1000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.204301075268817,
|
| 312 |
+
"grad_norm": 2.1475138664245605,
|
| 313 |
+
"learning_rate": 8.737349397590363e-06,
|
| 314 |
+
"loss": 0.2197,
|
| 315 |
+
"step": 1025
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 2.258064516129032,
|
| 319 |
+
"grad_norm": 2.0517914295196533,
|
| 320 |
+
"learning_rate": 8.67710843373494e-06,
|
| 321 |
+
"loss": 0.2045,
|
| 322 |
+
"step": 1050
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.3118279569892475,
|
| 326 |
+
"grad_norm": 2.6586711406707764,
|
| 327 |
+
"learning_rate": 8.616867469879518e-06,
|
| 328 |
+
"loss": 0.1677,
|
| 329 |
+
"step": 1075
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.3655913978494625,
|
| 333 |
+
"grad_norm": 2.79289174079895,
|
| 334 |
+
"learning_rate": 8.556626506024098e-06,
|
| 335 |
+
"loss": 0.1972,
|
| 336 |
+
"step": 1100
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.4193548387096775,
|
| 340 |
+
"grad_norm": 2.236177921295166,
|
| 341 |
+
"learning_rate": 8.496385542168676e-06,
|
| 342 |
+
"loss": 0.1762,
|
| 343 |
+
"step": 1125
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 2.4731182795698925,
|
| 347 |
+
"grad_norm": 2.2909016609191895,
|
| 348 |
+
"learning_rate": 8.436144578313253e-06,
|
| 349 |
+
"loss": 0.2063,
|
| 350 |
+
"step": 1150
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.5268817204301075,
|
| 354 |
+
"grad_norm": 2.6273820400238037,
|
| 355 |
+
"learning_rate": 8.375903614457833e-06,
|
| 356 |
+
"loss": 0.1859,
|
| 357 |
+
"step": 1175
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.5806451612903225,
|
| 361 |
+
"grad_norm": 1.9557993412017822,
|
| 362 |
+
"learning_rate": 8.31566265060241e-06,
|
| 363 |
+
"loss": 0.1887,
|
| 364 |
+
"step": 1200
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 2.6344086021505375,
|
| 368 |
+
"grad_norm": 2.2926785945892334,
|
| 369 |
+
"learning_rate": 8.255421686746988e-06,
|
| 370 |
+
"loss": 0.2071,
|
| 371 |
+
"step": 1225
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.688172043010753,
|
| 375 |
+
"grad_norm": 1.8259581327438354,
|
| 376 |
+
"learning_rate": 8.195180722891567e-06,
|
| 377 |
+
"loss": 0.1942,
|
| 378 |
+
"step": 1250
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 2.741935483870968,
|
| 382 |
+
"grad_norm": 2.699507713317871,
|
| 383 |
+
"learning_rate": 8.134939759036145e-06,
|
| 384 |
+
"loss": 0.2111,
|
| 385 |
+
"step": 1275
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 2.795698924731183,
|
| 389 |
+
"grad_norm": 2.465116262435913,
|
| 390 |
+
"learning_rate": 8.074698795180723e-06,
|
| 391 |
+
"loss": 0.1823,
|
| 392 |
+
"step": 1300
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 2.849462365591398,
|
| 396 |
+
"grad_norm": 2.7853310108184814,
|
| 397 |
+
"learning_rate": 8.014457831325302e-06,
|
| 398 |
+
"loss": 0.1874,
|
| 399 |
+
"step": 1325
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 2.903225806451613,
|
| 403 |
+
"grad_norm": 2.566476821899414,
|
| 404 |
+
"learning_rate": 7.95421686746988e-06,
|
| 405 |
+
"loss": 0.195,
|
| 406 |
+
"step": 1350
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 2.956989247311828,
|
| 410 |
+
"grad_norm": 2.295469045639038,
|
| 411 |
+
"learning_rate": 7.893975903614458e-06,
|
| 412 |
+
"loss": 0.1785,
|
| 413 |
+
"step": 1375
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.010752688172043,
|
| 417 |
+
"grad_norm": 1.3029838800430298,
|
| 418 |
+
"learning_rate": 7.833734939759037e-06,
|
| 419 |
+
"loss": 0.1552,
|
| 420 |
+
"step": 1400
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 3.064516129032258,
|
| 424 |
+
"grad_norm": 1.8084765672683716,
|
| 425 |
+
"learning_rate": 7.773493975903615e-06,
|
| 426 |
+
"loss": 0.1068,
|
| 427 |
+
"step": 1425
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 3.118279569892473,
|
| 431 |
+
"grad_norm": 1.5511784553527832,
|
| 432 |
+
"learning_rate": 7.713253012048193e-06,
|
| 433 |
+
"loss": 0.1118,
|
| 434 |
+
"step": 1450
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 3.172043010752688,
|
| 438 |
+
"grad_norm": 1.735387921333313,
|
| 439 |
+
"learning_rate": 7.653012048192772e-06,
|
| 440 |
+
"loss": 0.1034,
|
| 441 |
+
"step": 1475
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 3.225806451612903,
|
| 445 |
+
"grad_norm": 1.8845360279083252,
|
| 446 |
+
"learning_rate": 7.5927710843373505e-06,
|
| 447 |
+
"loss": 0.1076,
|
| 448 |
+
"step": 1500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 3.225806451612903,
|
| 452 |
+
"eval_loss": 0.30533668398857117,
|
| 453 |
+
"eval_runtime": 659.4803,
|
| 454 |
+
"eval_samples_per_second": 1.434,
|
| 455 |
+
"eval_steps_per_second": 1.434,
|
| 456 |
+
"eval_wer": 14.948691873902192,
|
| 457 |
+
"step": 1500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 3.279569892473118,
|
| 461 |
+
"grad_norm": 2.263770580291748,
|
| 462 |
+
"learning_rate": 7.532530120481928e-06,
|
| 463 |
+
"loss": 0.1056,
|
| 464 |
+
"step": 1525
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.3333333333333335,
|
| 468 |
+
"grad_norm": 2.537961483001709,
|
| 469 |
+
"learning_rate": 7.472289156626507e-06,
|
| 470 |
+
"loss": 0.1086,
|
| 471 |
+
"step": 1550
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.3870967741935485,
|
| 475 |
+
"grad_norm": 1.8475592136383057,
|
| 476 |
+
"learning_rate": 7.412048192771085e-06,
|
| 477 |
+
"loss": 0.1281,
|
| 478 |
+
"step": 1575
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 3.4408602150537635,
|
| 482 |
+
"grad_norm": 2.284437656402588,
|
| 483 |
+
"learning_rate": 7.351807228915663e-06,
|
| 484 |
+
"loss": 0.1106,
|
| 485 |
+
"step": 1600
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.4946236559139785,
|
| 489 |
+
"grad_norm": 2.063593626022339,
|
| 490 |
+
"learning_rate": 7.2915662650602415e-06,
|
| 491 |
+
"loss": 0.1154,
|
| 492 |
+
"step": 1625
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 3.5483870967741935,
|
| 496 |
+
"grad_norm": 2.439608573913574,
|
| 497 |
+
"learning_rate": 7.23132530120482e-06,
|
| 498 |
+
"loss": 0.1141,
|
| 499 |
+
"step": 1650
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 3.6021505376344085,
|
| 503 |
+
"grad_norm": 2.858224391937256,
|
| 504 |
+
"learning_rate": 7.171084337349398e-06,
|
| 505 |
+
"loss": 0.113,
|
| 506 |
+
"step": 1675
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 3.6559139784946235,
|
| 510 |
+
"grad_norm": 1.7737411260604858,
|
| 511 |
+
"learning_rate": 7.110843373493976e-06,
|
| 512 |
+
"loss": 0.1113,
|
| 513 |
+
"step": 1700
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.709677419354839,
|
| 517 |
+
"grad_norm": 2.2213852405548096,
|
| 518 |
+
"learning_rate": 7.050602409638555e-06,
|
| 519 |
+
"loss": 0.0834,
|
| 520 |
+
"step": 1725
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 3.763440860215054,
|
| 524 |
+
"grad_norm": 1.8823848962783813,
|
| 525 |
+
"learning_rate": 6.990361445783133e-06,
|
| 526 |
+
"loss": 0.0923,
|
| 527 |
+
"step": 1750
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 3.817204301075269,
|
| 531 |
+
"grad_norm": 1.8632618188858032,
|
| 532 |
+
"learning_rate": 6.930120481927711e-06,
|
| 533 |
+
"loss": 0.0973,
|
| 534 |
+
"step": 1775
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 3.870967741935484,
|
| 538 |
+
"grad_norm": 1.6536948680877686,
|
| 539 |
+
"learning_rate": 6.86987951807229e-06,
|
| 540 |
+
"loss": 0.1056,
|
| 541 |
+
"step": 1800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 3.924731182795699,
|
| 545 |
+
"grad_norm": 3.557621955871582,
|
| 546 |
+
"learning_rate": 6.809638554216867e-06,
|
| 547 |
+
"loss": 0.1222,
|
| 548 |
+
"step": 1825
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 3.978494623655914,
|
| 552 |
+
"grad_norm": 3.2628748416900635,
|
| 553 |
+
"learning_rate": 6.749397590361446e-06,
|
| 554 |
+
"loss": 0.1121,
|
| 555 |
+
"step": 1850
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 4.032258064516129,
|
| 559 |
+
"grad_norm": 1.7985060214996338,
|
| 560 |
+
"learning_rate": 6.6891566265060245e-06,
|
| 561 |
+
"loss": 0.0866,
|
| 562 |
+
"step": 1875
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 4.086021505376344,
|
| 566 |
+
"grad_norm": 2.224243402481079,
|
| 567 |
+
"learning_rate": 6.628915662650603e-06,
|
| 568 |
+
"loss": 0.0545,
|
| 569 |
+
"step": 1900
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 4.139784946236559,
|
| 573 |
+
"grad_norm": 1.2248103618621826,
|
| 574 |
+
"learning_rate": 6.568674698795181e-06,
|
| 575 |
+
"loss": 0.0496,
|
| 576 |
+
"step": 1925
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 4.193548387096774,
|
| 580 |
+
"grad_norm": 1.1298043727874756,
|
| 581 |
+
"learning_rate": 6.508433734939759e-06,
|
| 582 |
+
"loss": 0.0465,
|
| 583 |
+
"step": 1950
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 4.247311827956989,
|
| 587 |
+
"grad_norm": 1.406107783317566,
|
| 588 |
+
"learning_rate": 6.448192771084339e-06,
|
| 589 |
+
"loss": 0.0419,
|
| 590 |
+
"step": 1975
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 4.301075268817204,
|
| 594 |
+
"grad_norm": 1.4863886833190918,
|
| 595 |
+
"learning_rate": 6.387951807228916e-06,
|
| 596 |
+
"loss": 0.0493,
|
| 597 |
+
"step": 2000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 4.301075268817204,
|
| 601 |
+
"eval_loss": 0.3230736553668976,
|
| 602 |
+
"eval_runtime": 659.9391,
|
| 603 |
+
"eval_samples_per_second": 1.433,
|
| 604 |
+
"eval_steps_per_second": 1.433,
|
| 605 |
+
"eval_wer": 15.142830729407414,
|
| 606 |
+
"step": 2000
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 4.354838709677419,
|
| 610 |
+
"grad_norm": 1.9545135498046875,
|
| 611 |
+
"learning_rate": 6.327710843373495e-06,
|
| 612 |
+
"loss": 0.0561,
|
| 613 |
+
"step": 2025
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 4.408602150537634,
|
| 617 |
+
"grad_norm": 1.5624581575393677,
|
| 618 |
+
"learning_rate": 6.2674698795180735e-06,
|
| 619 |
+
"loss": 0.0507,
|
| 620 |
+
"step": 2050
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 4.462365591397849,
|
| 624 |
+
"grad_norm": 2.2645909786224365,
|
| 625 |
+
"learning_rate": 6.207228915662651e-06,
|
| 626 |
+
"loss": 0.0515,
|
| 627 |
+
"step": 2075
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 4.516129032258064,
|
| 631 |
+
"grad_norm": 1.7314592599868774,
|
| 632 |
+
"learning_rate": 6.14698795180723e-06,
|
| 633 |
+
"loss": 0.0592,
|
| 634 |
+
"step": 2100
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 4.56989247311828,
|
| 638 |
+
"grad_norm": 1.9412422180175781,
|
| 639 |
+
"learning_rate": 6.086746987951808e-06,
|
| 640 |
+
"loss": 0.0571,
|
| 641 |
+
"step": 2125
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 4.623655913978495,
|
| 645 |
+
"grad_norm": 1.7892636060714722,
|
| 646 |
+
"learning_rate": 6.026506024096386e-06,
|
| 647 |
+
"loss": 0.0523,
|
| 648 |
+
"step": 2150
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 4.67741935483871,
|
| 652 |
+
"grad_norm": 1.6095598936080933,
|
| 653 |
+
"learning_rate": 5.966265060240965e-06,
|
| 654 |
+
"loss": 0.0809,
|
| 655 |
+
"step": 2175
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 4.731182795698925,
|
| 659 |
+
"grad_norm": 1.4222418069839478,
|
| 660 |
+
"learning_rate": 5.906024096385543e-06,
|
| 661 |
+
"loss": 0.0447,
|
| 662 |
+
"step": 2200
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 4.78494623655914,
|
| 666 |
+
"grad_norm": 2.332413911819458,
|
| 667 |
+
"learning_rate": 5.845783132530121e-06,
|
| 668 |
+
"loss": 0.0597,
|
| 669 |
+
"step": 2225
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.838709677419355,
|
| 673 |
+
"grad_norm": 1.7514561414718628,
|
| 674 |
+
"learning_rate": 5.7855421686746994e-06,
|
| 675 |
+
"loss": 0.0553,
|
| 676 |
+
"step": 2250
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 4.89247311827957,
|
| 680 |
+
"grad_norm": 1.3207898139953613,
|
| 681 |
+
"learning_rate": 5.725301204819278e-06,
|
| 682 |
+
"loss": 0.0601,
|
| 683 |
+
"step": 2275
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 4.946236559139785,
|
| 687 |
+
"grad_norm": 1.5105242729187012,
|
| 688 |
+
"learning_rate": 5.665060240963856e-06,
|
| 689 |
+
"loss": 0.0459,
|
| 690 |
+
"step": 2300
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 5.0,
|
| 694 |
+
"grad_norm": 1.6449012756347656,
|
| 695 |
+
"learning_rate": 5.604819277108434e-06,
|
| 696 |
+
"loss": 0.0563,
|
| 697 |
+
"step": 2325
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 5.053763440860215,
|
| 701 |
+
"grad_norm": 0.5901522040367126,
|
| 702 |
+
"learning_rate": 5.544578313253013e-06,
|
| 703 |
+
"loss": 0.0223,
|
| 704 |
+
"step": 2350
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 5.10752688172043,
|
| 708 |
+
"grad_norm": 1.7701222896575928,
|
| 709 |
+
"learning_rate": 5.4843373493975905e-06,
|
| 710 |
+
"loss": 0.0297,
|
| 711 |
+
"step": 2375
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 5.161290322580645,
|
| 715 |
+
"grad_norm": 1.6865407228469849,
|
| 716 |
+
"learning_rate": 5.424096385542169e-06,
|
| 717 |
+
"loss": 0.025,
|
| 718 |
+
"step": 2400
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 5.21505376344086,
|
| 722 |
+
"grad_norm": 0.7481775283813477,
|
| 723 |
+
"learning_rate": 5.363855421686748e-06,
|
| 724 |
+
"loss": 0.028,
|
| 725 |
+
"step": 2425
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 5.268817204301075,
|
| 729 |
+
"grad_norm": 0.8327718377113342,
|
| 730 |
+
"learning_rate": 5.303614457831325e-06,
|
| 731 |
+
"loss": 0.0299,
|
| 732 |
+
"step": 2450
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 5.32258064516129,
|
| 736 |
+
"grad_norm": 0.9485449194908142,
|
| 737 |
+
"learning_rate": 5.243373493975904e-06,
|
| 738 |
+
"loss": 0.0344,
|
| 739 |
+
"step": 2475
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 5.376344086021505,
|
| 743 |
+
"grad_norm": 0.4464230239391327,
|
| 744 |
+
"learning_rate": 5.183132530120482e-06,
|
| 745 |
+
"loss": 0.0251,
|
| 746 |
+
"step": 2500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 5.376344086021505,
|
| 750 |
+
"eval_loss": 0.3420054018497467,
|
| 751 |
+
"eval_runtime": 660.0896,
|
| 752 |
+
"eval_samples_per_second": 1.433,
|
| 753 |
+
"eval_steps_per_second": 1.433,
|
| 754 |
+
"eval_wer": 14.560414162891744,
|
| 755 |
+
"step": 2500
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 5.43010752688172,
|
| 759 |
+
"grad_norm": 1.2224642038345337,
|
| 760 |
+
"learning_rate": 5.12289156626506e-06,
|
| 761 |
+
"loss": 0.0279,
|
| 762 |
+
"step": 2525
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 5.483870967741936,
|
| 766 |
+
"grad_norm": 1.1704018115997314,
|
| 767 |
+
"learning_rate": 5.062650602409639e-06,
|
| 768 |
+
"loss": 0.0268,
|
| 769 |
+
"step": 2550
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 5.53763440860215,
|
| 773 |
+
"grad_norm": 1.2785944938659668,
|
| 774 |
+
"learning_rate": 5.002409638554217e-06,
|
| 775 |
+
"loss": 0.024,
|
| 776 |
+
"step": 2575
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 5.591397849462366,
|
| 780 |
+
"grad_norm": 1.0779672861099243,
|
| 781 |
+
"learning_rate": 4.942168674698796e-06,
|
| 782 |
+
"loss": 0.0275,
|
| 783 |
+
"step": 2600
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 5.645161290322581,
|
| 787 |
+
"grad_norm": 1.1974660158157349,
|
| 788 |
+
"learning_rate": 4.8819277108433735e-06,
|
| 789 |
+
"loss": 0.0333,
|
| 790 |
+
"step": 2625
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 5.698924731182796,
|
| 794 |
+
"grad_norm": 1.610154151916504,
|
| 795 |
+
"learning_rate": 4.821686746987953e-06,
|
| 796 |
+
"loss": 0.0328,
|
| 797 |
+
"step": 2650
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 5.752688172043011,
|
| 801 |
+
"grad_norm": 1.3220410346984863,
|
| 802 |
+
"learning_rate": 4.761445783132531e-06,
|
| 803 |
+
"loss": 0.0271,
|
| 804 |
+
"step": 2675
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 5.806451612903226,
|
| 808 |
+
"grad_norm": 0.9111455082893372,
|
| 809 |
+
"learning_rate": 4.701204819277108e-06,
|
| 810 |
+
"loss": 0.0312,
|
| 811 |
+
"step": 2700
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 5.860215053763441,
|
| 815 |
+
"grad_norm": 0.685110867023468,
|
| 816 |
+
"learning_rate": 4.640963855421688e-06,
|
| 817 |
+
"loss": 0.0338,
|
| 818 |
+
"step": 2725
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 5.913978494623656,
|
| 822 |
+
"grad_norm": 1.8127559423446655,
|
| 823 |
+
"learning_rate": 4.580722891566265e-06,
|
| 824 |
+
"loss": 0.0243,
|
| 825 |
+
"step": 2750
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 5.967741935483871,
|
| 829 |
+
"grad_norm": 0.8406071662902832,
|
| 830 |
+
"learning_rate": 4.520481927710844e-06,
|
| 831 |
+
"loss": 0.0228,
|
| 832 |
+
"step": 2775
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 6.021505376344086,
|
| 836 |
+
"grad_norm": 1.7931848764419556,
|
| 837 |
+
"learning_rate": 4.4602409638554225e-06,
|
| 838 |
+
"loss": 0.0138,
|
| 839 |
+
"step": 2800
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 6.075268817204301,
|
| 843 |
+
"grad_norm": 0.9648796916007996,
|
| 844 |
+
"learning_rate": 4.4e-06,
|
| 845 |
+
"loss": 0.0145,
|
| 846 |
+
"step": 2825
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 6.129032258064516,
|
| 850 |
+
"grad_norm": 1.0061876773834229,
|
| 851 |
+
"learning_rate": 4.339759036144579e-06,
|
| 852 |
+
"loss": 0.0165,
|
| 853 |
+
"step": 2850
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 6.182795698924731,
|
| 857 |
+
"grad_norm": 1.3273780345916748,
|
| 858 |
+
"learning_rate": 4.279518072289157e-06,
|
| 859 |
+
"loss": 0.0217,
|
| 860 |
+
"step": 2875
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 6.236559139784946,
|
| 864 |
+
"grad_norm": 0.5284897089004517,
|
| 865 |
+
"learning_rate": 4.219277108433735e-06,
|
| 866 |
+
"loss": 0.0114,
|
| 867 |
+
"step": 2900
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 6.290322580645161,
|
| 871 |
+
"grad_norm": 0.27909165620803833,
|
| 872 |
+
"learning_rate": 4.159036144578314e-06,
|
| 873 |
+
"loss": 0.0142,
|
| 874 |
+
"step": 2925
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 6.344086021505376,
|
| 878 |
+
"grad_norm": 0.47097161412239075,
|
| 879 |
+
"learning_rate": 4.098795180722892e-06,
|
| 880 |
+
"loss": 0.0089,
|
| 881 |
+
"step": 2950
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 6.397849462365591,
|
| 885 |
+
"grad_norm": 0.6813268661499023,
|
| 886 |
+
"learning_rate": 4.03855421686747e-06,
|
| 887 |
+
"loss": 0.0176,
|
| 888 |
+
"step": 2975
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 6.451612903225806,
|
| 892 |
+
"grad_norm": 0.43314510583877563,
|
| 893 |
+
"learning_rate": 3.978313253012048e-06,
|
| 894 |
+
"loss": 0.0099,
|
| 895 |
+
"step": 3000
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 6.451612903225806,
|
| 899 |
+
"eval_loss": 0.35552269220352173,
|
| 900 |
+
"eval_runtime": 654.5926,
|
| 901 |
+
"eval_samples_per_second": 1.445,
|
| 902 |
+
"eval_steps_per_second": 1.445,
|
| 903 |
+
"eval_wer": 14.172136451881299,
|
| 904 |
+
"step": 3000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 6.505376344086022,
|
| 908 |
+
"grad_norm": 0.26192936301231384,
|
| 909 |
+
"learning_rate": 3.918072289156627e-06,
|
| 910 |
+
"loss": 0.0126,
|
| 911 |
+
"step": 3025
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 6.559139784946236,
|
| 915 |
+
"grad_norm": 1.1532106399536133,
|
| 916 |
+
"learning_rate": 3.857831325301205e-06,
|
| 917 |
+
"loss": 0.0176,
|
| 918 |
+
"step": 3050
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 6.612903225806452,
|
| 922 |
+
"grad_norm": 1.5923449993133545,
|
| 923 |
+
"learning_rate": 3.7975903614457832e-06,
|
| 924 |
+
"loss": 0.0152,
|
| 925 |
+
"step": 3075
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 6.666666666666667,
|
| 929 |
+
"grad_norm": 1.4764392375946045,
|
| 930 |
+
"learning_rate": 3.7373493975903618e-06,
|
| 931 |
+
"loss": 0.011,
|
| 932 |
+
"step": 3100
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 6.720430107526882,
|
| 936 |
+
"grad_norm": 0.7815900444984436,
|
| 937 |
+
"learning_rate": 3.67710843373494e-06,
|
| 938 |
+
"loss": 0.011,
|
| 939 |
+
"step": 3125
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 6.774193548387097,
|
| 943 |
+
"grad_norm": 0.5374385714530945,
|
| 944 |
+
"learning_rate": 3.6168674698795185e-06,
|
| 945 |
+
"loss": 0.0121,
|
| 946 |
+
"step": 3150
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 6.827956989247312,
|
| 950 |
+
"grad_norm": 0.8609434366226196,
|
| 951 |
+
"learning_rate": 3.5566265060240966e-06,
|
| 952 |
+
"loss": 0.0142,
|
| 953 |
+
"step": 3175
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 6.881720430107527,
|
| 957 |
+
"grad_norm": 1.7298405170440674,
|
| 958 |
+
"learning_rate": 3.4963855421686747e-06,
|
| 959 |
+
"loss": 0.0219,
|
| 960 |
+
"step": 3200
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 6.935483870967742,
|
| 964 |
+
"grad_norm": 1.2537510395050049,
|
| 965 |
+
"learning_rate": 3.4361445783132537e-06,
|
| 966 |
+
"loss": 0.0108,
|
| 967 |
+
"step": 3225
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 6.989247311827957,
|
| 971 |
+
"grad_norm": 1.078411340713501,
|
| 972 |
+
"learning_rate": 3.375903614457832e-06,
|
| 973 |
+
"loss": 0.0137,
|
| 974 |
+
"step": 3250
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 7.043010752688172,
|
| 978 |
+
"grad_norm": 1.0412614345550537,
|
| 979 |
+
"learning_rate": 3.31566265060241e-06,
|
| 980 |
+
"loss": 0.0109,
|
| 981 |
+
"step": 3275
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 7.096774193548387,
|
| 985 |
+
"grad_norm": 0.15878470242023468,
|
| 986 |
+
"learning_rate": 3.2554216867469885e-06,
|
| 987 |
+
"loss": 0.0066,
|
| 988 |
+
"step": 3300
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 7.150537634408602,
|
| 992 |
+
"grad_norm": 0.731324315071106,
|
| 993 |
+
"learning_rate": 3.1951807228915666e-06,
|
| 994 |
+
"loss": 0.0066,
|
| 995 |
+
"step": 3325
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 7.204301075268817,
|
| 999 |
+
"grad_norm": 0.6188691854476929,
|
| 1000 |
+
"learning_rate": 3.1349397590361448e-06,
|
| 1001 |
+
"loss": 0.0074,
|
| 1002 |
+
"step": 3350
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 7.258064516129032,
|
| 1006 |
+
"grad_norm": 0.3013724684715271,
|
| 1007 |
+
"learning_rate": 3.0746987951807233e-06,
|
| 1008 |
+
"loss": 0.0047,
|
| 1009 |
+
"step": 3375
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 7.311827956989247,
|
| 1013 |
+
"grad_norm": 0.20967040956020355,
|
| 1014 |
+
"learning_rate": 3.0144578313253014e-06,
|
| 1015 |
+
"loss": 0.0081,
|
| 1016 |
+
"step": 3400
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 7.365591397849462,
|
| 1020 |
+
"grad_norm": 0.5221308469772339,
|
| 1021 |
+
"learning_rate": 2.9542168674698796e-06,
|
| 1022 |
+
"loss": 0.0053,
|
| 1023 |
+
"step": 3425
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 7.419354838709677,
|
| 1027 |
+
"grad_norm": 1.011181116104126,
|
| 1028 |
+
"learning_rate": 2.893975903614458e-06,
|
| 1029 |
+
"loss": 0.0075,
|
| 1030 |
+
"step": 3450
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 7.473118279569892,
|
| 1034 |
+
"grad_norm": 0.49731409549713135,
|
| 1035 |
+
"learning_rate": 2.8337349397590363e-06,
|
| 1036 |
+
"loss": 0.0072,
|
| 1037 |
+
"step": 3475
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 7.526881720430108,
|
| 1041 |
+
"grad_norm": 0.6202843189239502,
|
| 1042 |
+
"learning_rate": 2.7734939759036144e-06,
|
| 1043 |
+
"loss": 0.0059,
|
| 1044 |
+
"step": 3500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 7.526881720430108,
|
| 1048 |
+
"eval_loss": 0.3649694621562958,
|
| 1049 |
+
"eval_runtime": 654.6952,
|
| 1050 |
+
"eval_samples_per_second": 1.445,
|
| 1051 |
+
"eval_steps_per_second": 1.445,
|
| 1052 |
+
"eval_wer": 14.107423500046224,
|
| 1053 |
+
"step": 3500
|
| 1054 |
+
}
|
| 1055 |
+
],
|
| 1056 |
+
"logging_steps": 25,
|
| 1057 |
+
"max_steps": 4650,
|
| 1058 |
+
"num_input_tokens_seen": 0,
|
| 1059 |
+
"num_train_epochs": 10,
|
| 1060 |
+
"save_steps": 500,
|
| 1061 |
+
"stateful_callbacks": {
|
| 1062 |
+
"TrainerControl": {
|
| 1063 |
+
"args": {
|
| 1064 |
+
"should_epoch_stop": false,
|
| 1065 |
+
"should_evaluate": false,
|
| 1066 |
+
"should_log": false,
|
| 1067 |
+
"should_save": true,
|
| 1068 |
+
"should_training_stop": false
|
| 1069 |
+
},
|
| 1070 |
+
"attributes": {}
|
| 1071 |
+
}
|
| 1072 |
+
},
|
| 1073 |
+
"total_flos": 3.8044846287028224e+20,
|
| 1074 |
+
"train_batch_size": 8,
|
| 1075 |
+
"trial_name": null,
|
| 1076 |
+
"trial_params": null
|
| 1077 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bb552be6aea78f3e838df39d34483577e7c3c649e24eef8816e3aae58d76742
|
| 3 |
+
size 5560
|