Upload folder using huggingface_hub
Browse files- config.json +61 -0
- generation_config.json +254 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- preprocessor_config.json +14 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +934 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "openai/whisper-small",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"apply_spec_augment": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"WhisperForConditionalGeneration"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"begin_suppress_tokens": null,
|
| 11 |
+
"bos_token_id": 50257,
|
| 12 |
+
"classifier_proj_size": 256,
|
| 13 |
+
"d_model": 768,
|
| 14 |
+
"decoder_attention_heads": 12,
|
| 15 |
+
"decoder_ffn_dim": 3072,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 12,
|
| 18 |
+
"decoder_start_token_id": 50258,
|
| 19 |
+
"dropout": 0.0,
|
| 20 |
+
"encoder_attention_heads": 12,
|
| 21 |
+
"encoder_ffn_dim": 3072,
|
| 22 |
+
"encoder_layerdrop": 0.0,
|
| 23 |
+
"encoder_layers": 12,
|
| 24 |
+
"eos_token_id": 50257,
|
| 25 |
+
"forced_decoder_ids": [
|
| 26 |
+
[
|
| 27 |
+
1,
|
| 28 |
+
50259
|
| 29 |
+
],
|
| 30 |
+
[
|
| 31 |
+
2,
|
| 32 |
+
50359
|
| 33 |
+
],
|
| 34 |
+
[
|
| 35 |
+
3,
|
| 36 |
+
50363
|
| 37 |
+
]
|
| 38 |
+
],
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"mask_feature_length": 10,
|
| 42 |
+
"mask_feature_min_masks": 0,
|
| 43 |
+
"mask_feature_prob": 0.0,
|
| 44 |
+
"mask_time_length": 10,
|
| 45 |
+
"mask_time_min_masks": 2,
|
| 46 |
+
"mask_time_prob": 0.05,
|
| 47 |
+
"max_length": null,
|
| 48 |
+
"max_source_positions": 1500,
|
| 49 |
+
"max_target_positions": 448,
|
| 50 |
+
"median_filter_width": 7,
|
| 51 |
+
"model_type": "whisper",
|
| 52 |
+
"num_hidden_layers": 12,
|
| 53 |
+
"num_mel_bins": 80,
|
| 54 |
+
"pad_token_id": 50257,
|
| 55 |
+
"scale_embedding": false,
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.48.1",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"use_weighted_layer_sum": false,
|
| 60 |
+
"vocab_size": 51865
|
| 61 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,254 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alignment_heads": [
|
| 3 |
+
[
|
| 4 |
+
5,
|
| 5 |
+
3
|
| 6 |
+
],
|
| 7 |
+
[
|
| 8 |
+
5,
|
| 9 |
+
9
|
| 10 |
+
],
|
| 11 |
+
[
|
| 12 |
+
8,
|
| 13 |
+
0
|
| 14 |
+
],
|
| 15 |
+
[
|
| 16 |
+
8,
|
| 17 |
+
4
|
| 18 |
+
],
|
| 19 |
+
[
|
| 20 |
+
8,
|
| 21 |
+
7
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
8,
|
| 25 |
+
8
|
| 26 |
+
],
|
| 27 |
+
[
|
| 28 |
+
9,
|
| 29 |
+
0
|
| 30 |
+
],
|
| 31 |
+
[
|
| 32 |
+
9,
|
| 33 |
+
7
|
| 34 |
+
],
|
| 35 |
+
[
|
| 36 |
+
9,
|
| 37 |
+
9
|
| 38 |
+
],
|
| 39 |
+
[
|
| 40 |
+
10,
|
| 41 |
+
5
|
| 42 |
+
]
|
| 43 |
+
],
|
| 44 |
+
"begin_suppress_tokens": [
|
| 45 |
+
220,
|
| 46 |
+
50257
|
| 47 |
+
],
|
| 48 |
+
"bos_token_id": 50257,
|
| 49 |
+
"decoder_start_token_id": 50258,
|
| 50 |
+
"eos_token_id": 50257,
|
| 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 |
+
"<|zh|>": 50260
|
| 152 |
+
},
|
| 153 |
+
"language": "fa",
|
| 154 |
+
"max_initial_timestamp_index": 50,
|
| 155 |
+
"max_length": 448,
|
| 156 |
+
"no_timestamps_token_id": 50363,
|
| 157 |
+
"pad_token_id": 50257,
|
| 158 |
+
"prev_sot_token_id": 50361,
|
| 159 |
+
"return_timestamps": false,
|
| 160 |
+
"suppress_tokens": [
|
| 161 |
+
1,
|
| 162 |
+
2,
|
| 163 |
+
7,
|
| 164 |
+
8,
|
| 165 |
+
9,
|
| 166 |
+
10,
|
| 167 |
+
14,
|
| 168 |
+
25,
|
| 169 |
+
26,
|
| 170 |
+
27,
|
| 171 |
+
28,
|
| 172 |
+
29,
|
| 173 |
+
31,
|
| 174 |
+
58,
|
| 175 |
+
59,
|
| 176 |
+
60,
|
| 177 |
+
61,
|
| 178 |
+
62,
|
| 179 |
+
63,
|
| 180 |
+
90,
|
| 181 |
+
91,
|
| 182 |
+
92,
|
| 183 |
+
93,
|
| 184 |
+
359,
|
| 185 |
+
503,
|
| 186 |
+
522,
|
| 187 |
+
542,
|
| 188 |
+
873,
|
| 189 |
+
893,
|
| 190 |
+
902,
|
| 191 |
+
918,
|
| 192 |
+
922,
|
| 193 |
+
931,
|
| 194 |
+
1350,
|
| 195 |
+
1853,
|
| 196 |
+
1982,
|
| 197 |
+
2460,
|
| 198 |
+
2627,
|
| 199 |
+
3246,
|
| 200 |
+
3253,
|
| 201 |
+
3268,
|
| 202 |
+
3536,
|
| 203 |
+
3846,
|
| 204 |
+
3961,
|
| 205 |
+
4183,
|
| 206 |
+
4667,
|
| 207 |
+
6585,
|
| 208 |
+
6647,
|
| 209 |
+
7273,
|
| 210 |
+
9061,
|
| 211 |
+
9383,
|
| 212 |
+
10428,
|
| 213 |
+
10929,
|
| 214 |
+
11938,
|
| 215 |
+
12033,
|
| 216 |
+
12331,
|
| 217 |
+
12562,
|
| 218 |
+
13793,
|
| 219 |
+
14157,
|
| 220 |
+
14635,
|
| 221 |
+
15265,
|
| 222 |
+
15618,
|
| 223 |
+
16553,
|
| 224 |
+
16604,
|
| 225 |
+
18362,
|
| 226 |
+
18956,
|
| 227 |
+
20075,
|
| 228 |
+
21675,
|
| 229 |
+
22520,
|
| 230 |
+
26130,
|
| 231 |
+
26161,
|
| 232 |
+
26435,
|
| 233 |
+
28279,
|
| 234 |
+
29464,
|
| 235 |
+
31650,
|
| 236 |
+
32302,
|
| 237 |
+
32470,
|
| 238 |
+
36865,
|
| 239 |
+
42863,
|
| 240 |
+
47425,
|
| 241 |
+
49870,
|
| 242 |
+
50254,
|
| 243 |
+
50258,
|
| 244 |
+
50360,
|
| 245 |
+
50361,
|
| 246 |
+
50362
|
| 247 |
+
],
|
| 248 |
+
"task": "transcribe",
|
| 249 |
+
"task_to_id": {
|
| 250 |
+
"transcribe": 50359,
|
| 251 |
+
"translate": 50358
|
| 252 |
+
},
|
| 253 |
+
"transformers_version": "4.48.1"
|
| 254 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59daa4f5ba5da7f7c29f188bd45683062e1e020e9aa83aa96ece84613687b172
|
| 3 |
+
size 966995080
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f89c8afe7ab26b700568570c5c80588fc085447dc08beb252138aa9624586af5
|
| 3 |
+
size 1925064044
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 4 |
+
"feature_size": 80,
|
| 5 |
+
"hop_length": 160,
|
| 6 |
+
"n_fft": 400,
|
| 7 |
+
"n_samples": 480000,
|
| 8 |
+
"nb_max_frames": 3000,
|
| 9 |
+
"padding_side": "right",
|
| 10 |
+
"padding_value": 0.0,
|
| 11 |
+
"processor_class": "WhisperProcessor",
|
| 12 |
+
"return_attention_mask": false,
|
| 13 |
+
"sampling_rate": 16000
|
| 14 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a5dde4942c5abf0a7e95dd68d15abde0ed07d4be0476227543b42039d647ca6
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:002588355b0a3efb0bf3c7df1a06a9309c26626dbba04576cc157176774fcec4
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,934 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 37.93360828750764,
|
| 3 |
+
"best_model_checkpoint": "/kaggle/working/whisper-small-fa/checkpoint-1200",
|
| 4 |
+
"epoch": 0.5862237420615535,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 1200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0004885197850512946,
|
| 13 |
+
"grad_norm": Infinity,
|
| 14 |
+
"learning_rate": 0.0,
|
| 15 |
+
"loss": 1.0812,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004885197850512946,
|
| 20 |
+
"grad_norm": 7.0439229011535645,
|
| 21 |
+
"learning_rate": 1.6e-07,
|
| 22 |
+
"loss": 1.1303,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.009770395701025891,
|
| 27 |
+
"grad_norm": 6.462230205535889,
|
| 28 |
+
"learning_rate": 3.6e-07,
|
| 29 |
+
"loss": 1.1685,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014655593551538837,
|
| 34 |
+
"grad_norm": 6.182502269744873,
|
| 35 |
+
"learning_rate": 5.6e-07,
|
| 36 |
+
"loss": 1.1928,
|
| 37 |
+
"step": 30
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.019540791402051783,
|
| 41 |
+
"grad_norm": 5.509102821350098,
|
| 42 |
+
"learning_rate": 7.6e-07,
|
| 43 |
+
"loss": 1.0699,
|
| 44 |
+
"step": 40
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.024425989252564728,
|
| 48 |
+
"grad_norm": 6.310844421386719,
|
| 49 |
+
"learning_rate": 9.600000000000001e-07,
|
| 50 |
+
"loss": 1.1117,
|
| 51 |
+
"step": 50
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.029311187103077674,
|
| 55 |
+
"grad_norm": 4.859102725982666,
|
| 56 |
+
"learning_rate": 1.1600000000000001e-06,
|
| 57 |
+
"loss": 1.0681,
|
| 58 |
+
"step": 60
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03419638495359062,
|
| 62 |
+
"grad_norm": 5.078814506530762,
|
| 63 |
+
"learning_rate": 1.3600000000000001e-06,
|
| 64 |
+
"loss": 1.0524,
|
| 65 |
+
"step": 70
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.039081582804103565,
|
| 69 |
+
"grad_norm": 5.108152866363525,
|
| 70 |
+
"learning_rate": 1.56e-06,
|
| 71 |
+
"loss": 0.9862,
|
| 72 |
+
"step": 80
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04396678065461651,
|
| 76 |
+
"grad_norm": 6.09797477722168,
|
| 77 |
+
"learning_rate": 1.76e-06,
|
| 78 |
+
"loss": 1.0003,
|
| 79 |
+
"step": 90
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.048851978505129456,
|
| 83 |
+
"grad_norm": 6.775981426239014,
|
| 84 |
+
"learning_rate": 1.9600000000000003e-06,
|
| 85 |
+
"loss": 0.9876,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0537371763556424,
|
| 90 |
+
"grad_norm": 4.734734535217285,
|
| 91 |
+
"learning_rate": 2.16e-06,
|
| 92 |
+
"loss": 0.9026,
|
| 93 |
+
"step": 110
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.05862237420615535,
|
| 97 |
+
"grad_norm": 5.119976043701172,
|
| 98 |
+
"learning_rate": 2.3600000000000003e-06,
|
| 99 |
+
"loss": 0.9783,
|
| 100 |
+
"step": 120
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0635075720566683,
|
| 104 |
+
"grad_norm": 4.984339714050293,
|
| 105 |
+
"learning_rate": 2.56e-06,
|
| 106 |
+
"loss": 0.9671,
|
| 107 |
+
"step": 130
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06839276990718124,
|
| 111 |
+
"grad_norm": 5.38236665725708,
|
| 112 |
+
"learning_rate": 2.7600000000000003e-06,
|
| 113 |
+
"loss": 0.9263,
|
| 114 |
+
"step": 140
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.07327796775769418,
|
| 118 |
+
"grad_norm": 4.587165355682373,
|
| 119 |
+
"learning_rate": 2.96e-06,
|
| 120 |
+
"loss": 0.9063,
|
| 121 |
+
"step": 150
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07816316560820713,
|
| 125 |
+
"grad_norm": 4.583350658416748,
|
| 126 |
+
"learning_rate": 3.1600000000000002e-06,
|
| 127 |
+
"loss": 0.84,
|
| 128 |
+
"step": 160
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.08304836345872008,
|
| 132 |
+
"grad_norm": 4.574127674102783,
|
| 133 |
+
"learning_rate": 3.3600000000000004e-06,
|
| 134 |
+
"loss": 0.833,
|
| 135 |
+
"step": 170
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.08793356130923302,
|
| 139 |
+
"grad_norm": 4.4891486167907715,
|
| 140 |
+
"learning_rate": 3.5600000000000002e-06,
|
| 141 |
+
"loss": 0.7965,
|
| 142 |
+
"step": 180
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.09281875915974597,
|
| 146 |
+
"grad_norm": 4.560169696807861,
|
| 147 |
+
"learning_rate": 3.7600000000000004e-06,
|
| 148 |
+
"loss": 0.7703,
|
| 149 |
+
"step": 190
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.09770395701025891,
|
| 153 |
+
"grad_norm": 5.059728145599365,
|
| 154 |
+
"learning_rate": 3.96e-06,
|
| 155 |
+
"loss": 0.8702,
|
| 156 |
+
"step": 200
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.09770395701025891,
|
| 160 |
+
"eval_loss": 0.7939865589141846,
|
| 161 |
+
"eval_runtime": 4279.5784,
|
| 162 |
+
"eval_samples_per_second": 0.956,
|
| 163 |
+
"eval_steps_per_second": 0.12,
|
| 164 |
+
"eval_wer": 55.69627284341499,
|
| 165 |
+
"step": 200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.10258915486077186,
|
| 169 |
+
"grad_norm": 4.930312156677246,
|
| 170 |
+
"learning_rate": 4.16e-06,
|
| 171 |
+
"loss": 0.7978,
|
| 172 |
+
"step": 210
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.1074743527112848,
|
| 176 |
+
"grad_norm": 4.10357666015625,
|
| 177 |
+
"learning_rate": 4.360000000000001e-06,
|
| 178 |
+
"loss": 0.7512,
|
| 179 |
+
"step": 220
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.11235955056179775,
|
| 183 |
+
"grad_norm": 4.250857830047607,
|
| 184 |
+
"learning_rate": 4.56e-06,
|
| 185 |
+
"loss": 0.7242,
|
| 186 |
+
"step": 230
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.1172447484123107,
|
| 190 |
+
"grad_norm": 4.367275714874268,
|
| 191 |
+
"learning_rate": 4.76e-06,
|
| 192 |
+
"loss": 0.7473,
|
| 193 |
+
"step": 240
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.12212994626282364,
|
| 197 |
+
"grad_norm": 5.22398042678833,
|
| 198 |
+
"learning_rate": 4.960000000000001e-06,
|
| 199 |
+
"loss": 0.8216,
|
| 200 |
+
"step": 250
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.1270151441133366,
|
| 204 |
+
"grad_norm": 4.164461135864258,
|
| 205 |
+
"learning_rate": 5.1600000000000006e-06,
|
| 206 |
+
"loss": 0.6435,
|
| 207 |
+
"step": 260
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.13190034196384953,
|
| 211 |
+
"grad_norm": 5.729864120483398,
|
| 212 |
+
"learning_rate": 5.36e-06,
|
| 213 |
+
"loss": 0.7428,
|
| 214 |
+
"step": 270
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.13678553981436248,
|
| 218 |
+
"grad_norm": 4.261190891265869,
|
| 219 |
+
"learning_rate": 5.560000000000001e-06,
|
| 220 |
+
"loss": 0.7081,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.14167073766487542,
|
| 225 |
+
"grad_norm": 3.940690279006958,
|
| 226 |
+
"learning_rate": 5.76e-06,
|
| 227 |
+
"loss": 0.63,
|
| 228 |
+
"step": 290
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.14655593551538837,
|
| 232 |
+
"grad_norm": 3.8406989574432373,
|
| 233 |
+
"learning_rate": 5.9600000000000005e-06,
|
| 234 |
+
"loss": 0.6711,
|
| 235 |
+
"step": 300
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.15144113336590131,
|
| 239 |
+
"grad_norm": 3.9562718868255615,
|
| 240 |
+
"learning_rate": 6.16e-06,
|
| 241 |
+
"loss": 0.6992,
|
| 242 |
+
"step": 310
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.15632633121641426,
|
| 246 |
+
"grad_norm": 4.3271589279174805,
|
| 247 |
+
"learning_rate": 6.360000000000001e-06,
|
| 248 |
+
"loss": 0.7054,
|
| 249 |
+
"step": 320
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.1612115290669272,
|
| 253 |
+
"grad_norm": 4.195059776306152,
|
| 254 |
+
"learning_rate": 6.560000000000001e-06,
|
| 255 |
+
"loss": 0.6424,
|
| 256 |
+
"step": 330
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.16609672691744015,
|
| 260 |
+
"grad_norm": 4.491285800933838,
|
| 261 |
+
"learning_rate": 6.760000000000001e-06,
|
| 262 |
+
"loss": 0.7025,
|
| 263 |
+
"step": 340
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.1709819247679531,
|
| 267 |
+
"grad_norm": 5.474099636077881,
|
| 268 |
+
"learning_rate": 6.96e-06,
|
| 269 |
+
"loss": 0.6373,
|
| 270 |
+
"step": 350
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.17586712261846604,
|
| 274 |
+
"grad_norm": 4.050487995147705,
|
| 275 |
+
"learning_rate": 7.16e-06,
|
| 276 |
+
"loss": 0.5873,
|
| 277 |
+
"step": 360
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 0.180752320468979,
|
| 281 |
+
"grad_norm": 5.067830562591553,
|
| 282 |
+
"learning_rate": 7.360000000000001e-06,
|
| 283 |
+
"loss": 0.6215,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.18563751831949193,
|
| 288 |
+
"grad_norm": 4.530464172363281,
|
| 289 |
+
"learning_rate": 7.5600000000000005e-06,
|
| 290 |
+
"loss": 0.6653,
|
| 291 |
+
"step": 380
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.19052271617000488,
|
| 295 |
+
"grad_norm": 3.94427490234375,
|
| 296 |
+
"learning_rate": 7.76e-06,
|
| 297 |
+
"loss": 0.6384,
|
| 298 |
+
"step": 390
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.19540791402051783,
|
| 302 |
+
"grad_norm": 4.769630432128906,
|
| 303 |
+
"learning_rate": 7.960000000000002e-06,
|
| 304 |
+
"loss": 0.6221,
|
| 305 |
+
"step": 400
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.19540791402051783,
|
| 309 |
+
"eval_loss": 0.6273319721221924,
|
| 310 |
+
"eval_runtime": 4292.0244,
|
| 311 |
+
"eval_samples_per_second": 0.954,
|
| 312 |
+
"eval_steps_per_second": 0.119,
|
| 313 |
+
"eval_wer": 46.51759428984058,
|
| 314 |
+
"step": 400
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.20029311187103077,
|
| 318 |
+
"grad_norm": 5.084413528442383,
|
| 319 |
+
"learning_rate": 8.16e-06,
|
| 320 |
+
"loss": 0.6848,
|
| 321 |
+
"step": 410
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.20517830972154372,
|
| 325 |
+
"grad_norm": 4.525864124298096,
|
| 326 |
+
"learning_rate": 8.36e-06,
|
| 327 |
+
"loss": 0.6234,
|
| 328 |
+
"step": 420
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.21006350757205666,
|
| 332 |
+
"grad_norm": 4.084338188171387,
|
| 333 |
+
"learning_rate": 8.560000000000001e-06,
|
| 334 |
+
"loss": 0.6272,
|
| 335 |
+
"step": 430
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.2149487054225696,
|
| 339 |
+
"grad_norm": 4.2073187828063965,
|
| 340 |
+
"learning_rate": 8.76e-06,
|
| 341 |
+
"loss": 0.6075,
|
| 342 |
+
"step": 440
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.21983390327308255,
|
| 346 |
+
"grad_norm": 3.7465310096740723,
|
| 347 |
+
"learning_rate": 8.96e-06,
|
| 348 |
+
"loss": 0.5911,
|
| 349 |
+
"step": 450
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.2247191011235955,
|
| 353 |
+
"grad_norm": 3.5266239643096924,
|
| 354 |
+
"learning_rate": 9.16e-06,
|
| 355 |
+
"loss": 0.5499,
|
| 356 |
+
"step": 460
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.22960429897410845,
|
| 360 |
+
"grad_norm": 4.592888832092285,
|
| 361 |
+
"learning_rate": 9.360000000000002e-06,
|
| 362 |
+
"loss": 0.6124,
|
| 363 |
+
"step": 470
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.2344894968246214,
|
| 367 |
+
"grad_norm": 4.9025726318359375,
|
| 368 |
+
"learning_rate": 9.56e-06,
|
| 369 |
+
"loss": 0.5697,
|
| 370 |
+
"step": 480
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.23937469467513434,
|
| 374 |
+
"grad_norm": 4.163031101226807,
|
| 375 |
+
"learning_rate": 9.760000000000001e-06,
|
| 376 |
+
"loss": 0.6106,
|
| 377 |
+
"step": 490
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.24425989252564728,
|
| 381 |
+
"grad_norm": 5.005125045776367,
|
| 382 |
+
"learning_rate": 9.960000000000001e-06,
|
| 383 |
+
"loss": 0.6138,
|
| 384 |
+
"step": 500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.24914509037616023,
|
| 388 |
+
"grad_norm": 3.767075538635254,
|
| 389 |
+
"learning_rate": 9.982222222222224e-06,
|
| 390 |
+
"loss": 0.601,
|
| 391 |
+
"step": 510
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.2540302882266732,
|
| 395 |
+
"grad_norm": 4.501799583435059,
|
| 396 |
+
"learning_rate": 9.960000000000001e-06,
|
| 397 |
+
"loss": 0.5664,
|
| 398 |
+
"step": 520
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.2589154860771861,
|
| 402 |
+
"grad_norm": 4.360989570617676,
|
| 403 |
+
"learning_rate": 9.937777777777779e-06,
|
| 404 |
+
"loss": 0.5782,
|
| 405 |
+
"step": 530
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.26380068392769906,
|
| 409 |
+
"grad_norm": 3.4221303462982178,
|
| 410 |
+
"learning_rate": 9.915555555555556e-06,
|
| 411 |
+
"loss": 0.585,
|
| 412 |
+
"step": 540
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 0.268685881778212,
|
| 416 |
+
"grad_norm": 4.4359540939331055,
|
| 417 |
+
"learning_rate": 9.893333333333334e-06,
|
| 418 |
+
"loss": 0.6203,
|
| 419 |
+
"step": 550
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.27357107962872496,
|
| 423 |
+
"grad_norm": 4.127051830291748,
|
| 424 |
+
"learning_rate": 9.871111111111112e-06,
|
| 425 |
+
"loss": 0.5032,
|
| 426 |
+
"step": 560
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.2784562774792379,
|
| 430 |
+
"grad_norm": 4.420245170593262,
|
| 431 |
+
"learning_rate": 9.84888888888889e-06,
|
| 432 |
+
"loss": 0.5203,
|
| 433 |
+
"step": 570
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.28334147532975085,
|
| 437 |
+
"grad_norm": 3.9797379970550537,
|
| 438 |
+
"learning_rate": 9.826666666666667e-06,
|
| 439 |
+
"loss": 0.5655,
|
| 440 |
+
"step": 580
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.2882266731802638,
|
| 444 |
+
"grad_norm": 4.368413925170898,
|
| 445 |
+
"learning_rate": 9.804444444444444e-06,
|
| 446 |
+
"loss": 0.5212,
|
| 447 |
+
"step": 590
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.29311187103077674,
|
| 451 |
+
"grad_norm": 4.136660099029541,
|
| 452 |
+
"learning_rate": 9.782222222222222e-06,
|
| 453 |
+
"loss": 0.5569,
|
| 454 |
+
"step": 600
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.29311187103077674,
|
| 458 |
+
"eval_loss": 0.5524213910102844,
|
| 459 |
+
"eval_runtime": 4286.0796,
|
| 460 |
+
"eval_samples_per_second": 0.955,
|
| 461 |
+
"eval_steps_per_second": 0.119,
|
| 462 |
+
"eval_wer": 43.251471976892745,
|
| 463 |
+
"step": 600
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.2979970688812897,
|
| 467 |
+
"grad_norm": 4.450418472290039,
|
| 468 |
+
"learning_rate": 9.760000000000001e-06,
|
| 469 |
+
"loss": 0.5764,
|
| 470 |
+
"step": 610
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.30288226673180263,
|
| 474 |
+
"grad_norm": 3.492920160293579,
|
| 475 |
+
"learning_rate": 9.737777777777779e-06,
|
| 476 |
+
"loss": 0.5373,
|
| 477 |
+
"step": 620
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.3077674645823156,
|
| 481 |
+
"grad_norm": 4.476955890655518,
|
| 482 |
+
"learning_rate": 9.715555555555557e-06,
|
| 483 |
+
"loss": 0.4991,
|
| 484 |
+
"step": 630
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.3126526624328285,
|
| 488 |
+
"grad_norm": 3.3835787773132324,
|
| 489 |
+
"learning_rate": 9.693333333333334e-06,
|
| 490 |
+
"loss": 0.5387,
|
| 491 |
+
"step": 640
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.31753786028334147,
|
| 495 |
+
"grad_norm": 3.3086416721343994,
|
| 496 |
+
"learning_rate": 9.671111111111112e-06,
|
| 497 |
+
"loss": 0.5414,
|
| 498 |
+
"step": 650
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.3224230581338544,
|
| 502 |
+
"grad_norm": 4.225744724273682,
|
| 503 |
+
"learning_rate": 9.64888888888889e-06,
|
| 504 |
+
"loss": 0.466,
|
| 505 |
+
"step": 660
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.32730825598436736,
|
| 509 |
+
"grad_norm": 4.349085807800293,
|
| 510 |
+
"learning_rate": 9.626666666666667e-06,
|
| 511 |
+
"loss": 0.5382,
|
| 512 |
+
"step": 670
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.3321934538348803,
|
| 516 |
+
"grad_norm": 3.8492863178253174,
|
| 517 |
+
"learning_rate": 9.604444444444445e-06,
|
| 518 |
+
"loss": 0.4948,
|
| 519 |
+
"step": 680
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.33707865168539325,
|
| 523 |
+
"grad_norm": 3.621649980545044,
|
| 524 |
+
"learning_rate": 9.582222222222222e-06,
|
| 525 |
+
"loss": 0.4967,
|
| 526 |
+
"step": 690
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.3419638495359062,
|
| 530 |
+
"grad_norm": 4.60444974899292,
|
| 531 |
+
"learning_rate": 9.56e-06,
|
| 532 |
+
"loss": 0.4764,
|
| 533 |
+
"step": 700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.34684904738641914,
|
| 537 |
+
"grad_norm": 3.711239814758301,
|
| 538 |
+
"learning_rate": 9.537777777777778e-06,
|
| 539 |
+
"loss": 0.5223,
|
| 540 |
+
"step": 710
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.3517342452369321,
|
| 544 |
+
"grad_norm": 3.4989352226257324,
|
| 545 |
+
"learning_rate": 9.515555555555557e-06,
|
| 546 |
+
"loss": 0.5236,
|
| 547 |
+
"step": 720
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.35661944308744503,
|
| 551 |
+
"grad_norm": 3.1489815711975098,
|
| 552 |
+
"learning_rate": 9.493333333333334e-06,
|
| 553 |
+
"loss": 0.5124,
|
| 554 |
+
"step": 730
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.361504640937958,
|
| 558 |
+
"grad_norm": 4.3361735343933105,
|
| 559 |
+
"learning_rate": 9.471111111111112e-06,
|
| 560 |
+
"loss": 0.5418,
|
| 561 |
+
"step": 740
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.3663898387884709,
|
| 565 |
+
"grad_norm": 2.7804179191589355,
|
| 566 |
+
"learning_rate": 9.44888888888889e-06,
|
| 567 |
+
"loss": 0.4497,
|
| 568 |
+
"step": 750
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.37127503663898387,
|
| 572 |
+
"grad_norm": 3.681450605392456,
|
| 573 |
+
"learning_rate": 9.426666666666667e-06,
|
| 574 |
+
"loss": 0.4977,
|
| 575 |
+
"step": 760
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.3761602344894968,
|
| 579 |
+
"grad_norm": 4.99979829788208,
|
| 580 |
+
"learning_rate": 9.404444444444445e-06,
|
| 581 |
+
"loss": 0.5545,
|
| 582 |
+
"step": 770
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 0.38104543234000976,
|
| 586 |
+
"grad_norm": 3.4073972702026367,
|
| 587 |
+
"learning_rate": 9.382222222222223e-06,
|
| 588 |
+
"loss": 0.4368,
|
| 589 |
+
"step": 780
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.3859306301905227,
|
| 593 |
+
"grad_norm": 3.558271646499634,
|
| 594 |
+
"learning_rate": 9.360000000000002e-06,
|
| 595 |
+
"loss": 0.4994,
|
| 596 |
+
"step": 790
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.39081582804103565,
|
| 600 |
+
"grad_norm": 3.8624231815338135,
|
| 601 |
+
"learning_rate": 9.33777777777778e-06,
|
| 602 |
+
"loss": 0.4785,
|
| 603 |
+
"step": 800
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.39081582804103565,
|
| 607 |
+
"eval_loss": 0.5109024047851562,
|
| 608 |
+
"eval_runtime": 4283.004,
|
| 609 |
+
"eval_samples_per_second": 0.956,
|
| 610 |
+
"eval_steps_per_second": 0.12,
|
| 611 |
+
"eval_wer": 40.329389546186746,
|
| 612 |
+
"step": 800
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3957010258915486,
|
| 616 |
+
"grad_norm": 5.084819793701172,
|
| 617 |
+
"learning_rate": 9.315555555555557e-06,
|
| 618 |
+
"loss": 0.5334,
|
| 619 |
+
"step": 810
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.40058622374206154,
|
| 623 |
+
"grad_norm": 4.143725872039795,
|
| 624 |
+
"learning_rate": 9.293333333333335e-06,
|
| 625 |
+
"loss": 0.4547,
|
| 626 |
+
"step": 820
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.4054714215925745,
|
| 630 |
+
"grad_norm": 4.0059614181518555,
|
| 631 |
+
"learning_rate": 9.271111111111112e-06,
|
| 632 |
+
"loss": 0.5135,
|
| 633 |
+
"step": 830
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.41035661944308743,
|
| 637 |
+
"grad_norm": 4.0563483238220215,
|
| 638 |
+
"learning_rate": 9.24888888888889e-06,
|
| 639 |
+
"loss": 0.4474,
|
| 640 |
+
"step": 840
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.4152418172936004,
|
| 644 |
+
"grad_norm": 3.5411062240600586,
|
| 645 |
+
"learning_rate": 9.226666666666668e-06,
|
| 646 |
+
"loss": 0.5204,
|
| 647 |
+
"step": 850
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.4201270151441133,
|
| 651 |
+
"grad_norm": 3.515293836593628,
|
| 652 |
+
"learning_rate": 9.204444444444445e-06,
|
| 653 |
+
"loss": 0.514,
|
| 654 |
+
"step": 860
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.42501221299462627,
|
| 658 |
+
"grad_norm": 3.598378896713257,
|
| 659 |
+
"learning_rate": 9.182222222222223e-06,
|
| 660 |
+
"loss": 0.4408,
|
| 661 |
+
"step": 870
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.4298974108451392,
|
| 665 |
+
"grad_norm": 3.1513373851776123,
|
| 666 |
+
"learning_rate": 9.16e-06,
|
| 667 |
+
"loss": 0.4575,
|
| 668 |
+
"step": 880
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.43478260869565216,
|
| 672 |
+
"grad_norm": 5.0970377922058105,
|
| 673 |
+
"learning_rate": 9.137777777777778e-06,
|
| 674 |
+
"loss": 0.4901,
|
| 675 |
+
"step": 890
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.4396678065461651,
|
| 679 |
+
"grad_norm": 4.093622207641602,
|
| 680 |
+
"learning_rate": 9.115555555555556e-06,
|
| 681 |
+
"loss": 0.4744,
|
| 682 |
+
"step": 900
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.44455300439667805,
|
| 686 |
+
"grad_norm": 4.782330513000488,
|
| 687 |
+
"learning_rate": 9.093333333333333e-06,
|
| 688 |
+
"loss": 0.4592,
|
| 689 |
+
"step": 910
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.449438202247191,
|
| 693 |
+
"grad_norm": 2.7775630950927734,
|
| 694 |
+
"learning_rate": 9.07111111111111e-06,
|
| 695 |
+
"loss": 0.511,
|
| 696 |
+
"step": 920
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.45432340009770394,
|
| 700 |
+
"grad_norm": 3.0217251777648926,
|
| 701 |
+
"learning_rate": 9.048888888888888e-06,
|
| 702 |
+
"loss": 0.4973,
|
| 703 |
+
"step": 930
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4592085979482169,
|
| 707 |
+
"grad_norm": 4.689064025878906,
|
| 708 |
+
"learning_rate": 9.026666666666666e-06,
|
| 709 |
+
"loss": 0.5676,
|
| 710 |
+
"step": 940
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.46409379579872984,
|
| 714 |
+
"grad_norm": 3.9128365516662598,
|
| 715 |
+
"learning_rate": 9.004444444444445e-06,
|
| 716 |
+
"loss": 0.4762,
|
| 717 |
+
"step": 950
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4689789936492428,
|
| 721 |
+
"grad_norm": 2.82621169090271,
|
| 722 |
+
"learning_rate": 8.982222222222223e-06,
|
| 723 |
+
"loss": 0.4318,
|
| 724 |
+
"step": 960
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.4738641914997557,
|
| 728 |
+
"grad_norm": 3.4691481590270996,
|
| 729 |
+
"learning_rate": 8.96e-06,
|
| 730 |
+
"loss": 0.4465,
|
| 731 |
+
"step": 970
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4787493893502687,
|
| 735 |
+
"grad_norm": 4.63604736328125,
|
| 736 |
+
"learning_rate": 8.937777777777778e-06,
|
| 737 |
+
"loss": 0.4895,
|
| 738 |
+
"step": 980
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.4836345872007816,
|
| 742 |
+
"grad_norm": 4.685360431671143,
|
| 743 |
+
"learning_rate": 8.915555555555556e-06,
|
| 744 |
+
"loss": 0.5155,
|
| 745 |
+
"step": 990
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.48851978505129456,
|
| 749 |
+
"grad_norm": 3.411714553833008,
|
| 750 |
+
"learning_rate": 8.893333333333333e-06,
|
| 751 |
+
"loss": 0.5216,
|
| 752 |
+
"step": 1000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.48851978505129456,
|
| 756 |
+
"eval_loss": 0.48320430517196655,
|
| 757 |
+
"eval_runtime": 4282.5384,
|
| 758 |
+
"eval_samples_per_second": 0.956,
|
| 759 |
+
"eval_steps_per_second": 0.12,
|
| 760 |
+
"eval_wer": 38.85151085930122,
|
| 761 |
+
"step": 1000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.4934049829018075,
|
| 765 |
+
"grad_norm": 3.4970908164978027,
|
| 766 |
+
"learning_rate": 8.871111111111111e-06,
|
| 767 |
+
"loss": 0.4371,
|
| 768 |
+
"step": 1010
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.49829018075232046,
|
| 772 |
+
"grad_norm": 4.287569522857666,
|
| 773 |
+
"learning_rate": 8.848888888888889e-06,
|
| 774 |
+
"loss": 0.4882,
|
| 775 |
+
"step": 1020
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.5031753786028335,
|
| 779 |
+
"grad_norm": 4.18141508102417,
|
| 780 |
+
"learning_rate": 8.826666666666668e-06,
|
| 781 |
+
"loss": 0.4955,
|
| 782 |
+
"step": 1030
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.5080605764533463,
|
| 786 |
+
"grad_norm": 3.6604855060577393,
|
| 787 |
+
"learning_rate": 8.804444444444446e-06,
|
| 788 |
+
"loss": 0.4448,
|
| 789 |
+
"step": 1040
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.5129457743038593,
|
| 793 |
+
"grad_norm": 4.121953010559082,
|
| 794 |
+
"learning_rate": 8.782222222222223e-06,
|
| 795 |
+
"loss": 0.5134,
|
| 796 |
+
"step": 1050
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.5178309721543722,
|
| 800 |
+
"grad_norm": 4.355337142944336,
|
| 801 |
+
"learning_rate": 8.76e-06,
|
| 802 |
+
"loss": 0.5069,
|
| 803 |
+
"step": 1060
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.5227161700048852,
|
| 807 |
+
"grad_norm": 3.3341116905212402,
|
| 808 |
+
"learning_rate": 8.737777777777778e-06,
|
| 809 |
+
"loss": 0.5008,
|
| 810 |
+
"step": 1070
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.5276013678553981,
|
| 814 |
+
"grad_norm": 4.42513370513916,
|
| 815 |
+
"learning_rate": 8.715555555555556e-06,
|
| 816 |
+
"loss": 0.4557,
|
| 817 |
+
"step": 1080
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.5324865657059111,
|
| 821 |
+
"grad_norm": 2.6197452545166016,
|
| 822 |
+
"learning_rate": 8.693333333333334e-06,
|
| 823 |
+
"loss": 0.458,
|
| 824 |
+
"step": 1090
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.537371763556424,
|
| 828 |
+
"grad_norm": 2.7881979942321777,
|
| 829 |
+
"learning_rate": 8.671111111111113e-06,
|
| 830 |
+
"loss": 0.4692,
|
| 831 |
+
"step": 1100
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.542256961406937,
|
| 835 |
+
"grad_norm": 3.419318914413452,
|
| 836 |
+
"learning_rate": 8.64888888888889e-06,
|
| 837 |
+
"loss": 0.4541,
|
| 838 |
+
"step": 1110
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.5471421592574499,
|
| 842 |
+
"grad_norm": 3.8821029663085938,
|
| 843 |
+
"learning_rate": 8.626666666666668e-06,
|
| 844 |
+
"loss": 0.5263,
|
| 845 |
+
"step": 1120
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.5520273571079629,
|
| 849 |
+
"grad_norm": 3.7360761165618896,
|
| 850 |
+
"learning_rate": 8.604444444444446e-06,
|
| 851 |
+
"loss": 0.479,
|
| 852 |
+
"step": 1130
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.5569125549584758,
|
| 856 |
+
"grad_norm": 3.3444161415100098,
|
| 857 |
+
"learning_rate": 8.582222222222223e-06,
|
| 858 |
+
"loss": 0.4942,
|
| 859 |
+
"step": 1140
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.5617977528089888,
|
| 863 |
+
"grad_norm": 3.6646621227264404,
|
| 864 |
+
"learning_rate": 8.560000000000001e-06,
|
| 865 |
+
"loss": 0.4129,
|
| 866 |
+
"step": 1150
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.5666829506595017,
|
| 870 |
+
"grad_norm": 3.4366016387939453,
|
| 871 |
+
"learning_rate": 8.537777777777779e-06,
|
| 872 |
+
"loss": 0.4324,
|
| 873 |
+
"step": 1160
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.5715681485100147,
|
| 877 |
+
"grad_norm": 3.347890615463257,
|
| 878 |
+
"learning_rate": 8.515555555555556e-06,
|
| 879 |
+
"loss": 0.4797,
|
| 880 |
+
"step": 1170
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.5764533463605276,
|
| 884 |
+
"grad_norm": 3.3374993801116943,
|
| 885 |
+
"learning_rate": 8.493333333333334e-06,
|
| 886 |
+
"loss": 0.4665,
|
| 887 |
+
"step": 1180
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.5813385442110406,
|
| 891 |
+
"grad_norm": 3.522141933441162,
|
| 892 |
+
"learning_rate": 8.471111111111112e-06,
|
| 893 |
+
"loss": 0.4141,
|
| 894 |
+
"step": 1190
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.5862237420615535,
|
| 898 |
+
"grad_norm": 3.1512022018432617,
|
| 899 |
+
"learning_rate": 8.448888888888889e-06,
|
| 900 |
+
"loss": 0.3981,
|
| 901 |
+
"step": 1200
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.5862237420615535,
|
| 905 |
+
"eval_loss": 0.4652426540851593,
|
| 906 |
+
"eval_runtime": 4278.5694,
|
| 907 |
+
"eval_samples_per_second": 0.957,
|
| 908 |
+
"eval_steps_per_second": 0.12,
|
| 909 |
+
"eval_wer": 37.93360828750764,
|
| 910 |
+
"step": 1200
|
| 911 |
+
}
|
| 912 |
+
],
|
| 913 |
+
"logging_steps": 10,
|
| 914 |
+
"max_steps": 5000,
|
| 915 |
+
"num_input_tokens_seen": 0,
|
| 916 |
+
"num_train_epochs": 3,
|
| 917 |
+
"save_steps": 200,
|
| 918 |
+
"stateful_callbacks": {
|
| 919 |
+
"TrainerControl": {
|
| 920 |
+
"args": {
|
| 921 |
+
"should_epoch_stop": false,
|
| 922 |
+
"should_evaluate": false,
|
| 923 |
+
"should_log": false,
|
| 924 |
+
"should_save": true,
|
| 925 |
+
"should_training_stop": false
|
| 926 |
+
},
|
| 927 |
+
"attributes": {}
|
| 928 |
+
}
|
| 929 |
+
},
|
| 930 |
+
"total_flos": 5.540839686144e+18,
|
| 931 |
+
"train_batch_size": 16,
|
| 932 |
+
"trial_name": null,
|
| 933 |
+
"trial_params": null
|
| 934 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43c7c7bf50493e1d1397926eca2fd6c036d324a5fceed54d64d1b6da0b45cf41
|
| 3 |
+
size 5432
|