Upload checkpoint directory
Browse files- checkpoint-5000/config.json +61 -0
- checkpoint-5000/generation_config.json +264 -0
- checkpoint-5000/model.safetensors +3 -0
- checkpoint-5000/optimizer.pt +3 -0
- checkpoint-5000/preprocessor_config.json +14 -0
- checkpoint-5000/rng_state.pth +3 -0
- checkpoint-5000/scheduler.pt +3 -0
- checkpoint-5000/trainer_state.json +1460 -0
- checkpoint-5000/training_args.bin +3 -0
- config.json +61 -0
- model.safetensors +3 -0
- preprocessor_config.json +14 -0
- training_args.bin +3 -0
checkpoint-5000/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Marwan-Kasem/whisper-small-hi42",
|
| 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": "4.47.0",
|
| 58 |
+
"use_cache": false,
|
| 59 |
+
"use_weighted_layer_sum": false,
|
| 60 |
+
"vocab_size": 51865
|
| 61 |
+
}
|
checkpoint-5000/generation_config.json
ADDED
|
@@ -0,0 +1,264 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"forced_decoder_ids": [
|
| 52 |
+
[
|
| 53 |
+
1,
|
| 54 |
+
null
|
| 55 |
+
],
|
| 56 |
+
[
|
| 57 |
+
2,
|
| 58 |
+
50359
|
| 59 |
+
]
|
| 60 |
+
],
|
| 61 |
+
"is_multilingual": true,
|
| 62 |
+
"lang_to_id": {
|
| 63 |
+
"<|af|>": 50327,
|
| 64 |
+
"<|am|>": 50334,
|
| 65 |
+
"<|ar|>": 50272,
|
| 66 |
+
"<|as|>": 50350,
|
| 67 |
+
"<|az|>": 50304,
|
| 68 |
+
"<|ba|>": 50355,
|
| 69 |
+
"<|be|>": 50330,
|
| 70 |
+
"<|bg|>": 50292,
|
| 71 |
+
"<|bn|>": 50302,
|
| 72 |
+
"<|bo|>": 50347,
|
| 73 |
+
"<|br|>": 50309,
|
| 74 |
+
"<|bs|>": 50315,
|
| 75 |
+
"<|ca|>": 50270,
|
| 76 |
+
"<|cs|>": 50283,
|
| 77 |
+
"<|cy|>": 50297,
|
| 78 |
+
"<|da|>": 50285,
|
| 79 |
+
"<|de|>": 50261,
|
| 80 |
+
"<|el|>": 50281,
|
| 81 |
+
"<|en|>": 50259,
|
| 82 |
+
"<|es|>": 50262,
|
| 83 |
+
"<|et|>": 50307,
|
| 84 |
+
"<|eu|>": 50310,
|
| 85 |
+
"<|fa|>": 50300,
|
| 86 |
+
"<|fi|>": 50277,
|
| 87 |
+
"<|fo|>": 50338,
|
| 88 |
+
"<|fr|>": 50265,
|
| 89 |
+
"<|gl|>": 50319,
|
| 90 |
+
"<|gu|>": 50333,
|
| 91 |
+
"<|haw|>": 50352,
|
| 92 |
+
"<|ha|>": 50354,
|
| 93 |
+
"<|he|>": 50279,
|
| 94 |
+
"<|hi|>": 50276,
|
| 95 |
+
"<|hr|>": 50291,
|
| 96 |
+
"<|ht|>": 50339,
|
| 97 |
+
"<|hu|>": 50286,
|
| 98 |
+
"<|hy|>": 50312,
|
| 99 |
+
"<|id|>": 50275,
|
| 100 |
+
"<|is|>": 50311,
|
| 101 |
+
"<|it|>": 50274,
|
| 102 |
+
"<|ja|>": 50266,
|
| 103 |
+
"<|jw|>": 50356,
|
| 104 |
+
"<|ka|>": 50329,
|
| 105 |
+
"<|kk|>": 50316,
|
| 106 |
+
"<|km|>": 50323,
|
| 107 |
+
"<|kn|>": 50306,
|
| 108 |
+
"<|ko|>": 50264,
|
| 109 |
+
"<|la|>": 50294,
|
| 110 |
+
"<|lb|>": 50345,
|
| 111 |
+
"<|ln|>": 50353,
|
| 112 |
+
"<|lo|>": 50336,
|
| 113 |
+
"<|lt|>": 50293,
|
| 114 |
+
"<|lv|>": 50301,
|
| 115 |
+
"<|mg|>": 50349,
|
| 116 |
+
"<|mi|>": 50295,
|
| 117 |
+
"<|mk|>": 50308,
|
| 118 |
+
"<|ml|>": 50296,
|
| 119 |
+
"<|mn|>": 50314,
|
| 120 |
+
"<|mr|>": 50320,
|
| 121 |
+
"<|ms|>": 50282,
|
| 122 |
+
"<|mt|>": 50343,
|
| 123 |
+
"<|my|>": 50346,
|
| 124 |
+
"<|ne|>": 50313,
|
| 125 |
+
"<|nl|>": 50271,
|
| 126 |
+
"<|nn|>": 50342,
|
| 127 |
+
"<|no|>": 50288,
|
| 128 |
+
"<|oc|>": 50328,
|
| 129 |
+
"<|pa|>": 50321,
|
| 130 |
+
"<|pl|>": 50269,
|
| 131 |
+
"<|ps|>": 50340,
|
| 132 |
+
"<|pt|>": 50267,
|
| 133 |
+
"<|ro|>": 50284,
|
| 134 |
+
"<|ru|>": 50263,
|
| 135 |
+
"<|sa|>": 50344,
|
| 136 |
+
"<|sd|>": 50332,
|
| 137 |
+
"<|si|>": 50322,
|
| 138 |
+
"<|sk|>": 50298,
|
| 139 |
+
"<|sl|>": 50305,
|
| 140 |
+
"<|sn|>": 50324,
|
| 141 |
+
"<|so|>": 50326,
|
| 142 |
+
"<|sq|>": 50317,
|
| 143 |
+
"<|sr|>": 50303,
|
| 144 |
+
"<|su|>": 50357,
|
| 145 |
+
"<|sv|>": 50273,
|
| 146 |
+
"<|sw|>": 50318,
|
| 147 |
+
"<|ta|>": 50287,
|
| 148 |
+
"<|te|>": 50299,
|
| 149 |
+
"<|tg|>": 50331,
|
| 150 |
+
"<|th|>": 50289,
|
| 151 |
+
"<|tk|>": 50341,
|
| 152 |
+
"<|tl|>": 50348,
|
| 153 |
+
"<|tr|>": 50268,
|
| 154 |
+
"<|tt|>": 50351,
|
| 155 |
+
"<|uk|>": 50280,
|
| 156 |
+
"<|ur|>": 50290,
|
| 157 |
+
"<|uz|>": 50337,
|
| 158 |
+
"<|vi|>": 50278,
|
| 159 |
+
"<|yi|>": 50335,
|
| 160 |
+
"<|yo|>": 50325,
|
| 161 |
+
"<|zh|>": 50260
|
| 162 |
+
},
|
| 163 |
+
"language": "english",
|
| 164 |
+
"max_initial_timestamp_index": 50,
|
| 165 |
+
"max_length": 448,
|
| 166 |
+
"no_timestamps_token_id": 50363,
|
| 167 |
+
"pad_token_id": 50257,
|
| 168 |
+
"prev_sot_token_id": 50361,
|
| 169 |
+
"return_timestamps": false,
|
| 170 |
+
"suppress_tokens": [
|
| 171 |
+
1,
|
| 172 |
+
2,
|
| 173 |
+
7,
|
| 174 |
+
8,
|
| 175 |
+
9,
|
| 176 |
+
10,
|
| 177 |
+
14,
|
| 178 |
+
25,
|
| 179 |
+
26,
|
| 180 |
+
27,
|
| 181 |
+
28,
|
| 182 |
+
29,
|
| 183 |
+
31,
|
| 184 |
+
58,
|
| 185 |
+
59,
|
| 186 |
+
60,
|
| 187 |
+
61,
|
| 188 |
+
62,
|
| 189 |
+
63,
|
| 190 |
+
90,
|
| 191 |
+
91,
|
| 192 |
+
92,
|
| 193 |
+
93,
|
| 194 |
+
359,
|
| 195 |
+
503,
|
| 196 |
+
522,
|
| 197 |
+
542,
|
| 198 |
+
873,
|
| 199 |
+
893,
|
| 200 |
+
902,
|
| 201 |
+
918,
|
| 202 |
+
922,
|
| 203 |
+
931,
|
| 204 |
+
1350,
|
| 205 |
+
1853,
|
| 206 |
+
1982,
|
| 207 |
+
2460,
|
| 208 |
+
2627,
|
| 209 |
+
3246,
|
| 210 |
+
3253,
|
| 211 |
+
3268,
|
| 212 |
+
3536,
|
| 213 |
+
3846,
|
| 214 |
+
3961,
|
| 215 |
+
4183,
|
| 216 |
+
4667,
|
| 217 |
+
6585,
|
| 218 |
+
6647,
|
| 219 |
+
7273,
|
| 220 |
+
9061,
|
| 221 |
+
9383,
|
| 222 |
+
10428,
|
| 223 |
+
10929,
|
| 224 |
+
11938,
|
| 225 |
+
12033,
|
| 226 |
+
12331,
|
| 227 |
+
12562,
|
| 228 |
+
13793,
|
| 229 |
+
14157,
|
| 230 |
+
14635,
|
| 231 |
+
15265,
|
| 232 |
+
15618,
|
| 233 |
+
16553,
|
| 234 |
+
16604,
|
| 235 |
+
18362,
|
| 236 |
+
18956,
|
| 237 |
+
20075,
|
| 238 |
+
21675,
|
| 239 |
+
22520,
|
| 240 |
+
26130,
|
| 241 |
+
26161,
|
| 242 |
+
26435,
|
| 243 |
+
28279,
|
| 244 |
+
29464,
|
| 245 |
+
31650,
|
| 246 |
+
32302,
|
| 247 |
+
32470,
|
| 248 |
+
36865,
|
| 249 |
+
42863,
|
| 250 |
+
47425,
|
| 251 |
+
49870,
|
| 252 |
+
50254,
|
| 253 |
+
50258,
|
| 254 |
+
50360,
|
| 255 |
+
50361,
|
| 256 |
+
50362
|
| 257 |
+
],
|
| 258 |
+
"task": "transcribe",
|
| 259 |
+
"task_to_id": {
|
| 260 |
+
"transcribe": 50359,
|
| 261 |
+
"translate": 50358
|
| 262 |
+
},
|
| 263 |
+
"transformers_version": "4.47.0"
|
| 264 |
+
}
|
checkpoint-5000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7725fd97010404929ac3190fe96450acde6e0fa5c682bc1d797fd6445141392c
|
| 3 |
+
size 966995080
|
checkpoint-5000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:27dd860ce33287935aec8fd2313d2c76aaf70f18c7608b87ee534539d825d041
|
| 3 |
+
size 1925064044
|
checkpoint-5000/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 |
+
}
|
checkpoint-5000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc95685584141f48b03398083068ebb84d7ea0b8823c3b4c31a52f8ee87aedf6
|
| 3 |
+
size 14244
|
checkpoint-5000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95fccff3d542d6c16277c65e5812a6d56ba8bfd572ee45d3b997123c549b8ec2
|
| 3 |
+
size 1064
|
checkpoint-5000/trainer_state.json
ADDED
|
@@ -0,0 +1,1460 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 894.1176470588235,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-hi42/checkpoint-2500",
|
| 4 |
+
"epoch": 0.737300007373,
|
| 5 |
+
"eval_steps": 2500,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0036865000368650003,
|
| 13 |
+
"grad_norm": 18.75660514831543,
|
| 14 |
+
"learning_rate": 1.26e-06,
|
| 15 |
+
"loss": 0.2281,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007373000073730001,
|
| 20 |
+
"grad_norm": 9.051491737365723,
|
| 21 |
+
"learning_rate": 2.76e-06,
|
| 22 |
+
"loss": 0.2149,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.011059500110595002,
|
| 27 |
+
"grad_norm": 18.25735092163086,
|
| 28 |
+
"learning_rate": 4.26e-06,
|
| 29 |
+
"loss": 0.1847,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014746000147460001,
|
| 34 |
+
"grad_norm": 15.795555114746094,
|
| 35 |
+
"learning_rate": 5.76e-06,
|
| 36 |
+
"loss": 0.1578,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.018432500184325,
|
| 41 |
+
"grad_norm": 19.176794052124023,
|
| 42 |
+
"learning_rate": 7.26e-06,
|
| 43 |
+
"loss": 0.1898,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.022119000221190004,
|
| 48 |
+
"grad_norm": 12.162376403808594,
|
| 49 |
+
"learning_rate": 8.759999999999999e-06,
|
| 50 |
+
"loss": 0.1925,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025805500258055003,
|
| 55 |
+
"grad_norm": 21.789493560791016,
|
| 56 |
+
"learning_rate": 1.0260000000000002e-05,
|
| 57 |
+
"loss": 0.1932,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.029492000294920002,
|
| 62 |
+
"grad_norm": 19.523834228515625,
|
| 63 |
+
"learning_rate": 1.1760000000000001e-05,
|
| 64 |
+
"loss": 0.2111,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.033178500331785,
|
| 69 |
+
"grad_norm": 29.254791259765625,
|
| 70 |
+
"learning_rate": 1.326e-05,
|
| 71 |
+
"loss": 0.181,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03686500036865,
|
| 76 |
+
"grad_norm": 24.781688690185547,
|
| 77 |
+
"learning_rate": 1.4760000000000001e-05,
|
| 78 |
+
"loss": 0.2074,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.040551500405515,
|
| 83 |
+
"grad_norm": 81.7341079711914,
|
| 84 |
+
"learning_rate": 1.626e-05,
|
| 85 |
+
"loss": 0.263,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04423800044238001,
|
| 90 |
+
"grad_norm": 15.537775993347168,
|
| 91 |
+
"learning_rate": 1.776e-05,
|
| 92 |
+
"loss": 0.2319,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04792450047924501,
|
| 97 |
+
"grad_norm": 30.62382698059082,
|
| 98 |
+
"learning_rate": 1.9260000000000002e-05,
|
| 99 |
+
"loss": 0.2183,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.051611000516110006,
|
| 104 |
+
"grad_norm": 17.783288955688477,
|
| 105 |
+
"learning_rate": 2.0759999999999998e-05,
|
| 106 |
+
"loss": 0.2514,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.055297500552975005,
|
| 111 |
+
"grad_norm": 17.249370574951172,
|
| 112 |
+
"learning_rate": 2.226e-05,
|
| 113 |
+
"loss": 0.2235,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.058984000589840005,
|
| 118 |
+
"grad_norm": 15.030986785888672,
|
| 119 |
+
"learning_rate": 2.3760000000000003e-05,
|
| 120 |
+
"loss": 0.2287,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06267050062670501,
|
| 125 |
+
"grad_norm": 13.615246772766113,
|
| 126 |
+
"learning_rate": 2.526e-05,
|
| 127 |
+
"loss": 0.2055,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06635700066357,
|
| 132 |
+
"grad_norm": 16.132198333740234,
|
| 133 |
+
"learning_rate": 2.676e-05,
|
| 134 |
+
"loss": 0.3079,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07004350070043501,
|
| 139 |
+
"grad_norm": 14.509906768798828,
|
| 140 |
+
"learning_rate": 2.826e-05,
|
| 141 |
+
"loss": 0.2644,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0737300007373,
|
| 146 |
+
"grad_norm": 26.63431167602539,
|
| 147 |
+
"learning_rate": 2.976e-05,
|
| 148 |
+
"loss": 0.2891,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07741650077416501,
|
| 153 |
+
"grad_norm": 19.87978744506836,
|
| 154 |
+
"learning_rate": 2.986e-05,
|
| 155 |
+
"loss": 0.303,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08110300081103,
|
| 160 |
+
"grad_norm": 20.629268646240234,
|
| 161 |
+
"learning_rate": 2.9693333333333333e-05,
|
| 162 |
+
"loss": 0.276,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08478950084789501,
|
| 167 |
+
"grad_norm": 21.41944122314453,
|
| 168 |
+
"learning_rate": 2.9526666666666666e-05,
|
| 169 |
+
"loss": 0.3054,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08847600088476001,
|
| 174 |
+
"grad_norm": 22.30516815185547,
|
| 175 |
+
"learning_rate": 2.936e-05,
|
| 176 |
+
"loss": 0.2826,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.092162500921625,
|
| 181 |
+
"grad_norm": 26.738222122192383,
|
| 182 |
+
"learning_rate": 2.9193333333333334e-05,
|
| 183 |
+
"loss": 0.2567,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09584900095849001,
|
| 188 |
+
"grad_norm": 20.48783302307129,
|
| 189 |
+
"learning_rate": 2.9026666666666664e-05,
|
| 190 |
+
"loss": 0.2818,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.099535500995355,
|
| 195 |
+
"grad_norm": 20.21194839477539,
|
| 196 |
+
"learning_rate": 2.8859999999999998e-05,
|
| 197 |
+
"loss": 0.2581,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10322200103222001,
|
| 202 |
+
"grad_norm": 17.05075454711914,
|
| 203 |
+
"learning_rate": 2.8693333333333335e-05,
|
| 204 |
+
"loss": 0.3104,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.106908501069085,
|
| 209 |
+
"grad_norm": 21.780668258666992,
|
| 210 |
+
"learning_rate": 2.852666666666667e-05,
|
| 211 |
+
"loss": 0.2956,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11059500110595001,
|
| 216 |
+
"grad_norm": 68.6823959350586,
|
| 217 |
+
"learning_rate": 2.8360000000000003e-05,
|
| 218 |
+
"loss": 0.24,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11428150114281502,
|
| 223 |
+
"grad_norm": 11.34193229675293,
|
| 224 |
+
"learning_rate": 2.8193333333333337e-05,
|
| 225 |
+
"loss": 0.2695,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11796800117968001,
|
| 230 |
+
"grad_norm": 16.80828285217285,
|
| 231 |
+
"learning_rate": 2.8026666666666667e-05,
|
| 232 |
+
"loss": 0.2308,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12165450121654502,
|
| 237 |
+
"grad_norm": 12.456579208374023,
|
| 238 |
+
"learning_rate": 2.786e-05,
|
| 239 |
+
"loss": 0.2881,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12534100125341002,
|
| 244 |
+
"grad_norm": 23.235368728637695,
|
| 245 |
+
"learning_rate": 2.7693333333333335e-05,
|
| 246 |
+
"loss": 0.2695,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12902750129027502,
|
| 251 |
+
"grad_norm": 38.54967498779297,
|
| 252 |
+
"learning_rate": 2.752666666666667e-05,
|
| 253 |
+
"loss": 0.292,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13271400132714,
|
| 258 |
+
"grad_norm": 26.577716827392578,
|
| 259 |
+
"learning_rate": 2.7360000000000002e-05,
|
| 260 |
+
"loss": 0.2879,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.136400501364005,
|
| 265 |
+
"grad_norm": 21.612504959106445,
|
| 266 |
+
"learning_rate": 2.7193333333333336e-05,
|
| 267 |
+
"loss": 0.2637,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.14008700140087002,
|
| 272 |
+
"grad_norm": 19.746545791625977,
|
| 273 |
+
"learning_rate": 2.7026666666666667e-05,
|
| 274 |
+
"loss": 0.2681,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.143773501437735,
|
| 279 |
+
"grad_norm": 24.593490600585938,
|
| 280 |
+
"learning_rate": 2.686e-05,
|
| 281 |
+
"loss": 0.2613,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1474600014746,
|
| 286 |
+
"grad_norm": 16.352689743041992,
|
| 287 |
+
"learning_rate": 2.6700000000000002e-05,
|
| 288 |
+
"loss": 0.3057,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.15114650151146503,
|
| 293 |
+
"grad_norm": 13.311758041381836,
|
| 294 |
+
"learning_rate": 2.6533333333333336e-05,
|
| 295 |
+
"loss": 0.2774,
|
| 296 |
+
"step": 1025
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15483300154833002,
|
| 300 |
+
"grad_norm": 17.516489028930664,
|
| 301 |
+
"learning_rate": 2.636666666666667e-05,
|
| 302 |
+
"loss": 0.2854,
|
| 303 |
+
"step": 1050
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.158519501585195,
|
| 307 |
+
"grad_norm": 29.11515998840332,
|
| 308 |
+
"learning_rate": 2.62e-05,
|
| 309 |
+
"loss": 0.2443,
|
| 310 |
+
"step": 1075
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16220600162206,
|
| 314 |
+
"grad_norm": 22.366535186767578,
|
| 315 |
+
"learning_rate": 2.6033333333333334e-05,
|
| 316 |
+
"loss": 0.269,
|
| 317 |
+
"step": 1100
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16589250165892502,
|
| 321 |
+
"grad_norm": 22.265789031982422,
|
| 322 |
+
"learning_rate": 2.5866666666666667e-05,
|
| 323 |
+
"loss": 0.2491,
|
| 324 |
+
"step": 1125
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16957900169579002,
|
| 328 |
+
"grad_norm": 27.416053771972656,
|
| 329 |
+
"learning_rate": 2.57e-05,
|
| 330 |
+
"loss": 0.298,
|
| 331 |
+
"step": 1150
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.173265501732655,
|
| 335 |
+
"grad_norm": 46.90145492553711,
|
| 336 |
+
"learning_rate": 2.5533333333333335e-05,
|
| 337 |
+
"loss": 0.2701,
|
| 338 |
+
"step": 1175
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17695200176952003,
|
| 342 |
+
"grad_norm": 17.11941146850586,
|
| 343 |
+
"learning_rate": 2.5366666666666665e-05,
|
| 344 |
+
"loss": 0.277,
|
| 345 |
+
"step": 1200
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.18063850180638502,
|
| 349 |
+
"grad_norm": 18.886322021484375,
|
| 350 |
+
"learning_rate": 2.52e-05,
|
| 351 |
+
"loss": 0.286,
|
| 352 |
+
"step": 1225
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18432500184325,
|
| 356 |
+
"grad_norm": 31.968734741210938,
|
| 357 |
+
"learning_rate": 2.504e-05,
|
| 358 |
+
"loss": 0.3451,
|
| 359 |
+
"step": 1250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.188011501880115,
|
| 363 |
+
"grad_norm": 21.282339096069336,
|
| 364 |
+
"learning_rate": 2.4873333333333335e-05,
|
| 365 |
+
"loss": 0.2694,
|
| 366 |
+
"step": 1275
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.19169800191698003,
|
| 370 |
+
"grad_norm": 20.881519317626953,
|
| 371 |
+
"learning_rate": 2.4713333333333336e-05,
|
| 372 |
+
"loss": 0.2769,
|
| 373 |
+
"step": 1300
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19538450195384502,
|
| 377 |
+
"grad_norm": 26.259872436523438,
|
| 378 |
+
"learning_rate": 2.4546666666666667e-05,
|
| 379 |
+
"loss": 0.2351,
|
| 380 |
+
"step": 1325
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19907100199071,
|
| 384 |
+
"grad_norm": 20.44563865661621,
|
| 385 |
+
"learning_rate": 2.438e-05,
|
| 386 |
+
"loss": 0.2618,
|
| 387 |
+
"step": 1350
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.20275750202757503,
|
| 391 |
+
"grad_norm": 20.123743057250977,
|
| 392 |
+
"learning_rate": 2.4213333333333334e-05,
|
| 393 |
+
"loss": 0.2596,
|
| 394 |
+
"step": 1375
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.20644400206444002,
|
| 398 |
+
"grad_norm": 20.72513198852539,
|
| 399 |
+
"learning_rate": 2.4046666666666668e-05,
|
| 400 |
+
"loss": 0.3285,
|
| 401 |
+
"step": 1400
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.21013050210130502,
|
| 405 |
+
"grad_norm": 15.452094078063965,
|
| 406 |
+
"learning_rate": 2.3880000000000002e-05,
|
| 407 |
+
"loss": 0.2632,
|
| 408 |
+
"step": 1425
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.21381700213817,
|
| 412 |
+
"grad_norm": 28.110380172729492,
|
| 413 |
+
"learning_rate": 2.3713333333333332e-05,
|
| 414 |
+
"loss": 0.3212,
|
| 415 |
+
"step": 1450
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21750350217503503,
|
| 419 |
+
"grad_norm": 29.40423583984375,
|
| 420 |
+
"learning_rate": 2.3546666666666666e-05,
|
| 421 |
+
"loss": 0.2967,
|
| 422 |
+
"step": 1475
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.22119000221190002,
|
| 426 |
+
"grad_norm": 18.681171417236328,
|
| 427 |
+
"learning_rate": 2.338e-05,
|
| 428 |
+
"loss": 0.2611,
|
| 429 |
+
"step": 1500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.22487650224876501,
|
| 433 |
+
"grad_norm": 20.94416046142578,
|
| 434 |
+
"learning_rate": 2.3213333333333334e-05,
|
| 435 |
+
"loss": 0.2489,
|
| 436 |
+
"step": 1525
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22856300228563003,
|
| 440 |
+
"grad_norm": 20.738452911376953,
|
| 441 |
+
"learning_rate": 2.3046666666666667e-05,
|
| 442 |
+
"loss": 0.232,
|
| 443 |
+
"step": 1550
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.23224950232249503,
|
| 447 |
+
"grad_norm": 20.05936050415039,
|
| 448 |
+
"learning_rate": 2.288e-05,
|
| 449 |
+
"loss": 0.3029,
|
| 450 |
+
"step": 1575
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23593600235936002,
|
| 454 |
+
"grad_norm": 20.57016944885254,
|
| 455 |
+
"learning_rate": 2.271333333333333e-05,
|
| 456 |
+
"loss": 0.2726,
|
| 457 |
+
"step": 1600
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.239622502396225,
|
| 461 |
+
"grad_norm": 26.36762237548828,
|
| 462 |
+
"learning_rate": 2.2546666666666665e-05,
|
| 463 |
+
"loss": 0.3293,
|
| 464 |
+
"step": 1625
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.24330900243309003,
|
| 468 |
+
"grad_norm": 15.222780227661133,
|
| 469 |
+
"learning_rate": 2.238e-05,
|
| 470 |
+
"loss": 0.2739,
|
| 471 |
+
"step": 1650
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24699550246995503,
|
| 475 |
+
"grad_norm": 14.14052963256836,
|
| 476 |
+
"learning_rate": 2.2213333333333333e-05,
|
| 477 |
+
"loss": 0.3381,
|
| 478 |
+
"step": 1675
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.25068200250682005,
|
| 482 |
+
"grad_norm": 24.071794509887695,
|
| 483 |
+
"learning_rate": 2.2046666666666667e-05,
|
| 484 |
+
"loss": 0.2647,
|
| 485 |
+
"step": 1700
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.25436850254368504,
|
| 489 |
+
"grad_norm": 17.39927101135254,
|
| 490 |
+
"learning_rate": 2.188e-05,
|
| 491 |
+
"loss": 0.2568,
|
| 492 |
+
"step": 1725
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.25805500258055003,
|
| 496 |
+
"grad_norm": 10.502955436706543,
|
| 497 |
+
"learning_rate": 2.1713333333333334e-05,
|
| 498 |
+
"loss": 0.2547,
|
| 499 |
+
"step": 1750
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.261741502617415,
|
| 503 |
+
"grad_norm": 10.240669250488281,
|
| 504 |
+
"learning_rate": 2.1546666666666668e-05,
|
| 505 |
+
"loss": 0.2874,
|
| 506 |
+
"step": 1775
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.26542800265428,
|
| 510 |
+
"grad_norm": 14.164012908935547,
|
| 511 |
+
"learning_rate": 2.1380000000000002e-05,
|
| 512 |
+
"loss": 0.2762,
|
| 513 |
+
"step": 1800
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.269114502691145,
|
| 517 |
+
"grad_norm": 20.989524841308594,
|
| 518 |
+
"learning_rate": 2.1213333333333336e-05,
|
| 519 |
+
"loss": 0.2981,
|
| 520 |
+
"step": 1825
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.27280100272801,
|
| 524 |
+
"grad_norm": 24.458831787109375,
|
| 525 |
+
"learning_rate": 2.104666666666667e-05,
|
| 526 |
+
"loss": 0.2788,
|
| 527 |
+
"step": 1850
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.27648750276487505,
|
| 531 |
+
"grad_norm": 14.625513076782227,
|
| 532 |
+
"learning_rate": 2.088e-05,
|
| 533 |
+
"loss": 0.252,
|
| 534 |
+
"step": 1875
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.28017400280174004,
|
| 538 |
+
"grad_norm": 19.73975944519043,
|
| 539 |
+
"learning_rate": 2.0713333333333334e-05,
|
| 540 |
+
"loss": 0.288,
|
| 541 |
+
"step": 1900
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.28386050283860503,
|
| 545 |
+
"grad_norm": 13.620248794555664,
|
| 546 |
+
"learning_rate": 2.0546666666666668e-05,
|
| 547 |
+
"loss": 0.2949,
|
| 548 |
+
"step": 1925
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.28754700287547,
|
| 552 |
+
"grad_norm": 30.92563247680664,
|
| 553 |
+
"learning_rate": 2.038e-05,
|
| 554 |
+
"loss": 0.2783,
|
| 555 |
+
"step": 1950
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.291233502912335,
|
| 559 |
+
"grad_norm": 23.79368782043457,
|
| 560 |
+
"learning_rate": 2.0213333333333335e-05,
|
| 561 |
+
"loss": 0.2906,
|
| 562 |
+
"step": 1975
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2949200029492,
|
| 566 |
+
"grad_norm": 20.57227897644043,
|
| 567 |
+
"learning_rate": 2.004666666666667e-05,
|
| 568 |
+
"loss": 0.3046,
|
| 569 |
+
"step": 2000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.298606502986065,
|
| 573 |
+
"grad_norm": 15.093461036682129,
|
| 574 |
+
"learning_rate": 1.988e-05,
|
| 575 |
+
"loss": 0.3137,
|
| 576 |
+
"step": 2025
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.30229300302293005,
|
| 580 |
+
"grad_norm": 20.4903564453125,
|
| 581 |
+
"learning_rate": 1.9713333333333333e-05,
|
| 582 |
+
"loss": 0.2995,
|
| 583 |
+
"step": 2050
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.30597950305979504,
|
| 587 |
+
"grad_norm": 15.112691879272461,
|
| 588 |
+
"learning_rate": 1.9546666666666667e-05,
|
| 589 |
+
"loss": 0.2575,
|
| 590 |
+
"step": 2075
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.30966600309666004,
|
| 594 |
+
"grad_norm": 54.09142303466797,
|
| 595 |
+
"learning_rate": 1.938e-05,
|
| 596 |
+
"loss": 0.2476,
|
| 597 |
+
"step": 2100
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.31335250313352503,
|
| 601 |
+
"grad_norm": 17.167198181152344,
|
| 602 |
+
"learning_rate": 1.9213333333333335e-05,
|
| 603 |
+
"loss": 0.2134,
|
| 604 |
+
"step": 2125
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.31703900317039,
|
| 608 |
+
"grad_norm": 27.618459701538086,
|
| 609 |
+
"learning_rate": 1.9046666666666665e-05,
|
| 610 |
+
"loss": 0.284,
|
| 611 |
+
"step": 2150
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.320725503207255,
|
| 615 |
+
"grad_norm": 17.521888732910156,
|
| 616 |
+
"learning_rate": 1.888e-05,
|
| 617 |
+
"loss": 0.2772,
|
| 618 |
+
"step": 2175
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.32441200324412,
|
| 622 |
+
"grad_norm": 21.149431228637695,
|
| 623 |
+
"learning_rate": 1.8713333333333333e-05,
|
| 624 |
+
"loss": 0.2058,
|
| 625 |
+
"step": 2200
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.32809850328098505,
|
| 629 |
+
"grad_norm": 17.894245147705078,
|
| 630 |
+
"learning_rate": 1.8546666666666666e-05,
|
| 631 |
+
"loss": 0.2566,
|
| 632 |
+
"step": 2225
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.33178500331785005,
|
| 636 |
+
"grad_norm": 17.015249252319336,
|
| 637 |
+
"learning_rate": 1.838e-05,
|
| 638 |
+
"loss": 0.2207,
|
| 639 |
+
"step": 2250
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.33547150335471504,
|
| 643 |
+
"grad_norm": 11.300625801086426,
|
| 644 |
+
"learning_rate": 1.8213333333333334e-05,
|
| 645 |
+
"loss": 0.2179,
|
| 646 |
+
"step": 2275
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.33915800339158003,
|
| 650 |
+
"grad_norm": 19.790618896484375,
|
| 651 |
+
"learning_rate": 1.8046666666666664e-05,
|
| 652 |
+
"loss": 0.2918,
|
| 653 |
+
"step": 2300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.342844503428445,
|
| 657 |
+
"grad_norm": 18.35441017150879,
|
| 658 |
+
"learning_rate": 1.7879999999999998e-05,
|
| 659 |
+
"loss": 0.2519,
|
| 660 |
+
"step": 2325
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.34653100346531,
|
| 664 |
+
"grad_norm": 23.775989532470703,
|
| 665 |
+
"learning_rate": 1.7713333333333332e-05,
|
| 666 |
+
"loss": 0.2713,
|
| 667 |
+
"step": 2350
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.350217503502175,
|
| 671 |
+
"grad_norm": 18.242185592651367,
|
| 672 |
+
"learning_rate": 1.754666666666667e-05,
|
| 673 |
+
"loss": 0.2145,
|
| 674 |
+
"step": 2375
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.35390400353904006,
|
| 678 |
+
"grad_norm": 26.26616096496582,
|
| 679 |
+
"learning_rate": 1.7380000000000003e-05,
|
| 680 |
+
"loss": 0.2913,
|
| 681 |
+
"step": 2400
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.35759050357590505,
|
| 685 |
+
"grad_norm": 18.030847549438477,
|
| 686 |
+
"learning_rate": 1.7213333333333337e-05,
|
| 687 |
+
"loss": 0.2446,
|
| 688 |
+
"step": 2425
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.36127700361277004,
|
| 692 |
+
"grad_norm": 14.601178169250488,
|
| 693 |
+
"learning_rate": 1.7046666666666667e-05,
|
| 694 |
+
"loss": 0.269,
|
| 695 |
+
"step": 2450
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.36496350364963503,
|
| 699 |
+
"grad_norm": 19.659399032592773,
|
| 700 |
+
"learning_rate": 1.688e-05,
|
| 701 |
+
"loss": 0.2094,
|
| 702 |
+
"step": 2475
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3686500036865,
|
| 706 |
+
"grad_norm": 11.19994068145752,
|
| 707 |
+
"learning_rate": 1.6713333333333335e-05,
|
| 708 |
+
"loss": 0.2417,
|
| 709 |
+
"step": 2500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3686500036865,
|
| 713 |
+
"eval_loss": 3.362957239151001,
|
| 714 |
+
"eval_runtime": 6743.3326,
|
| 715 |
+
"eval_samples_per_second": 0.938,
|
| 716 |
+
"eval_steps_per_second": 0.234,
|
| 717 |
+
"eval_wer": 894.1176470588235,
|
| 718 |
+
"step": 2500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.372336503723365,
|
| 722 |
+
"grad_norm": 16.059974670410156,
|
| 723 |
+
"learning_rate": 1.654666666666667e-05,
|
| 724 |
+
"loss": 0.2334,
|
| 725 |
+
"step": 2525
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.37602300376023,
|
| 729 |
+
"grad_norm": 14.174128532409668,
|
| 730 |
+
"learning_rate": 1.6380000000000002e-05,
|
| 731 |
+
"loss": 0.2674,
|
| 732 |
+
"step": 2550
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.37970950379709506,
|
| 736 |
+
"grad_norm": 17.93998908996582,
|
| 737 |
+
"learning_rate": 1.6213333333333333e-05,
|
| 738 |
+
"loss": 0.2318,
|
| 739 |
+
"step": 2575
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.38339600383396005,
|
| 743 |
+
"grad_norm": 19.523021697998047,
|
| 744 |
+
"learning_rate": 1.6046666666666667e-05,
|
| 745 |
+
"loss": 0.2847,
|
| 746 |
+
"step": 2600
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.38708250387082505,
|
| 750 |
+
"grad_norm": 16.256864547729492,
|
| 751 |
+
"learning_rate": 1.588e-05,
|
| 752 |
+
"loss": 0.2454,
|
| 753 |
+
"step": 2625
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.39076900390769004,
|
| 757 |
+
"grad_norm": 16.150251388549805,
|
| 758 |
+
"learning_rate": 1.5713333333333334e-05,
|
| 759 |
+
"loss": 0.305,
|
| 760 |
+
"step": 2650
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.39445550394455503,
|
| 764 |
+
"grad_norm": 21.875558853149414,
|
| 765 |
+
"learning_rate": 1.5546666666666668e-05,
|
| 766 |
+
"loss": 0.2634,
|
| 767 |
+
"step": 2675
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.39814200398142,
|
| 771 |
+
"grad_norm": 30.196002960205078,
|
| 772 |
+
"learning_rate": 1.5380000000000002e-05,
|
| 773 |
+
"loss": 0.2776,
|
| 774 |
+
"step": 2700
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.401828504018285,
|
| 778 |
+
"grad_norm": 22.096120834350586,
|
| 779 |
+
"learning_rate": 1.5213333333333332e-05,
|
| 780 |
+
"loss": 0.2298,
|
| 781 |
+
"step": 2725
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.40551500405515006,
|
| 785 |
+
"grad_norm": 17.02240753173828,
|
| 786 |
+
"learning_rate": 1.5046666666666666e-05,
|
| 787 |
+
"loss": 0.2348,
|
| 788 |
+
"step": 2750
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.40920150409201506,
|
| 792 |
+
"grad_norm": 19.941190719604492,
|
| 793 |
+
"learning_rate": 1.488e-05,
|
| 794 |
+
"loss": 0.2744,
|
| 795 |
+
"step": 2775
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.41288800412888005,
|
| 799 |
+
"grad_norm": 22.23070526123047,
|
| 800 |
+
"learning_rate": 1.4713333333333334e-05,
|
| 801 |
+
"loss": 0.2258,
|
| 802 |
+
"step": 2800
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.41657450416574504,
|
| 806 |
+
"grad_norm": 26.95673942565918,
|
| 807 |
+
"learning_rate": 1.4546666666666666e-05,
|
| 808 |
+
"loss": 0.2613,
|
| 809 |
+
"step": 2825
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.42026100420261003,
|
| 813 |
+
"grad_norm": 20.80782699584961,
|
| 814 |
+
"learning_rate": 1.438e-05,
|
| 815 |
+
"loss": 0.2921,
|
| 816 |
+
"step": 2850
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.423947504239475,
|
| 820 |
+
"grad_norm": 24.43097496032715,
|
| 821 |
+
"learning_rate": 1.4213333333333335e-05,
|
| 822 |
+
"loss": 0.2622,
|
| 823 |
+
"step": 2875
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.42763400427634,
|
| 827 |
+
"grad_norm": 17.876245498657227,
|
| 828 |
+
"learning_rate": 1.4046666666666667e-05,
|
| 829 |
+
"loss": 0.1954,
|
| 830 |
+
"step": 2900
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.43132050431320507,
|
| 834 |
+
"grad_norm": 47.99320602416992,
|
| 835 |
+
"learning_rate": 1.3880000000000001e-05,
|
| 836 |
+
"loss": 0.2499,
|
| 837 |
+
"step": 2925
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.43500700435007006,
|
| 841 |
+
"grad_norm": 15.234955787658691,
|
| 842 |
+
"learning_rate": 1.3713333333333335e-05,
|
| 843 |
+
"loss": 0.2234,
|
| 844 |
+
"step": 2950
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.43869350438693505,
|
| 848 |
+
"grad_norm": 15.011786460876465,
|
| 849 |
+
"learning_rate": 1.3546666666666667e-05,
|
| 850 |
+
"loss": 0.2197,
|
| 851 |
+
"step": 2975
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.44238000442380004,
|
| 855 |
+
"grad_norm": 11.17838191986084,
|
| 856 |
+
"learning_rate": 1.338e-05,
|
| 857 |
+
"loss": 0.2727,
|
| 858 |
+
"step": 3000
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.44606650446066504,
|
| 862 |
+
"grad_norm": 25.07973861694336,
|
| 863 |
+
"learning_rate": 1.3213333333333334e-05,
|
| 864 |
+
"loss": 0.2727,
|
| 865 |
+
"step": 3025
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.44975300449753003,
|
| 869 |
+
"grad_norm": 26.32314682006836,
|
| 870 |
+
"learning_rate": 1.3046666666666667e-05,
|
| 871 |
+
"loss": 0.2603,
|
| 872 |
+
"step": 3050
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.453439504534395,
|
| 876 |
+
"grad_norm": 16.592912673950195,
|
| 877 |
+
"learning_rate": 1.288e-05,
|
| 878 |
+
"loss": 0.2207,
|
| 879 |
+
"step": 3075
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.45712600457126007,
|
| 883 |
+
"grad_norm": 18.071935653686523,
|
| 884 |
+
"learning_rate": 1.2713333333333332e-05,
|
| 885 |
+
"loss": 0.2798,
|
| 886 |
+
"step": 3100
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.46081250460812506,
|
| 890 |
+
"grad_norm": 22.99540138244629,
|
| 891 |
+
"learning_rate": 1.2546666666666666e-05,
|
| 892 |
+
"loss": 0.2311,
|
| 893 |
+
"step": 3125
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 0.46449900464499005,
|
| 897 |
+
"grad_norm": 22.59821891784668,
|
| 898 |
+
"learning_rate": 1.238e-05,
|
| 899 |
+
"loss": 0.2279,
|
| 900 |
+
"step": 3150
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 0.46818550468185505,
|
| 904 |
+
"grad_norm": 18.768529891967773,
|
| 905 |
+
"learning_rate": 1.2213333333333334e-05,
|
| 906 |
+
"loss": 0.2425,
|
| 907 |
+
"step": 3175
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.47187200471872004,
|
| 911 |
+
"grad_norm": 15.747322082519531,
|
| 912 |
+
"learning_rate": 1.2046666666666668e-05,
|
| 913 |
+
"loss": 0.2469,
|
| 914 |
+
"step": 3200
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.47555850475558503,
|
| 918 |
+
"grad_norm": 17.282495498657227,
|
| 919 |
+
"learning_rate": 1.1880000000000001e-05,
|
| 920 |
+
"loss": 0.234,
|
| 921 |
+
"step": 3225
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.47924500479245,
|
| 925 |
+
"grad_norm": 16.40440559387207,
|
| 926 |
+
"learning_rate": 1.1713333333333334e-05,
|
| 927 |
+
"loss": 0.2485,
|
| 928 |
+
"step": 3250
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.4829315048293151,
|
| 932 |
+
"grad_norm": 21.604019165039062,
|
| 933 |
+
"learning_rate": 1.1546666666666667e-05,
|
| 934 |
+
"loss": 0.2254,
|
| 935 |
+
"step": 3275
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.48661800486618007,
|
| 939 |
+
"grad_norm": 13.803950309753418,
|
| 940 |
+
"learning_rate": 1.1380000000000001e-05,
|
| 941 |
+
"loss": 0.2021,
|
| 942 |
+
"step": 3300
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.49030450490304506,
|
| 946 |
+
"grad_norm": 24.17011833190918,
|
| 947 |
+
"learning_rate": 1.1213333333333333e-05,
|
| 948 |
+
"loss": 0.2255,
|
| 949 |
+
"step": 3325
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 0.49399100493991005,
|
| 953 |
+
"grad_norm": 15.46372127532959,
|
| 954 |
+
"learning_rate": 1.1046666666666667e-05,
|
| 955 |
+
"loss": 0.2457,
|
| 956 |
+
"step": 3350
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.49767750497677504,
|
| 960 |
+
"grad_norm": 18.39287757873535,
|
| 961 |
+
"learning_rate": 1.0880000000000001e-05,
|
| 962 |
+
"loss": 0.247,
|
| 963 |
+
"step": 3375
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.5013640050136401,
|
| 967 |
+
"grad_norm": 11.330999374389648,
|
| 968 |
+
"learning_rate": 1.0713333333333333e-05,
|
| 969 |
+
"loss": 0.2107,
|
| 970 |
+
"step": 3400
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.5050505050505051,
|
| 974 |
+
"grad_norm": 28.71451759338379,
|
| 975 |
+
"learning_rate": 1.0546666666666667e-05,
|
| 976 |
+
"loss": 0.2347,
|
| 977 |
+
"step": 3425
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.5087370050873701,
|
| 981 |
+
"grad_norm": 15.915485382080078,
|
| 982 |
+
"learning_rate": 1.0379999999999999e-05,
|
| 983 |
+
"loss": 0.2187,
|
| 984 |
+
"step": 3450
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.5124235051242351,
|
| 988 |
+
"grad_norm": 15.072096824645996,
|
| 989 |
+
"learning_rate": 1.0213333333333333e-05,
|
| 990 |
+
"loss": 0.2176,
|
| 991 |
+
"step": 3475
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.5161100051611001,
|
| 995 |
+
"grad_norm": 10.299458503723145,
|
| 996 |
+
"learning_rate": 1.0046666666666668e-05,
|
| 997 |
+
"loss": 0.3022,
|
| 998 |
+
"step": 3500
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.519796505197965,
|
| 1002 |
+
"grad_norm": 21.137348175048828,
|
| 1003 |
+
"learning_rate": 9.88e-06,
|
| 1004 |
+
"loss": 0.228,
|
| 1005 |
+
"step": 3525
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.52348300523483,
|
| 1009 |
+
"grad_norm": 6.446236610412598,
|
| 1010 |
+
"learning_rate": 9.713333333333334e-06,
|
| 1011 |
+
"loss": 0.2073,
|
| 1012 |
+
"step": 3550
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.527169505271695,
|
| 1016 |
+
"grad_norm": 15.32019329071045,
|
| 1017 |
+
"learning_rate": 9.546666666666668e-06,
|
| 1018 |
+
"loss": 0.2021,
|
| 1019 |
+
"step": 3575
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 0.53085600530856,
|
| 1023 |
+
"grad_norm": 20.475473403930664,
|
| 1024 |
+
"learning_rate": 9.38e-06,
|
| 1025 |
+
"loss": 0.2035,
|
| 1026 |
+
"step": 3600
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.534542505345425,
|
| 1030 |
+
"grad_norm": 11.656693458557129,
|
| 1031 |
+
"learning_rate": 9.213333333333334e-06,
|
| 1032 |
+
"loss": 0.2024,
|
| 1033 |
+
"step": 3625
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.53822900538229,
|
| 1037 |
+
"grad_norm": 15.80279541015625,
|
| 1038 |
+
"learning_rate": 9.046666666666668e-06,
|
| 1039 |
+
"loss": 0.2721,
|
| 1040 |
+
"step": 3650
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.541915505419155,
|
| 1044 |
+
"grad_norm": 14.904827117919922,
|
| 1045 |
+
"learning_rate": 8.88e-06,
|
| 1046 |
+
"loss": 0.199,
|
| 1047 |
+
"step": 3675
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 0.54560200545602,
|
| 1051 |
+
"grad_norm": 31.738067626953125,
|
| 1052 |
+
"learning_rate": 8.713333333333333e-06,
|
| 1053 |
+
"loss": 0.2582,
|
| 1054 |
+
"step": 3700
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 0.5492885054928851,
|
| 1058 |
+
"grad_norm": 16.204824447631836,
|
| 1059 |
+
"learning_rate": 8.546666666666667e-06,
|
| 1060 |
+
"loss": 0.2278,
|
| 1061 |
+
"step": 3725
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 0.5529750055297501,
|
| 1065 |
+
"grad_norm": 12.920398712158203,
|
| 1066 |
+
"learning_rate": 8.38e-06,
|
| 1067 |
+
"loss": 0.1984,
|
| 1068 |
+
"step": 3750
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 0.5566615055666151,
|
| 1072 |
+
"grad_norm": 11.068310737609863,
|
| 1073 |
+
"learning_rate": 8.213333333333333e-06,
|
| 1074 |
+
"loss": 0.2222,
|
| 1075 |
+
"step": 3775
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.5603480056034801,
|
| 1079 |
+
"grad_norm": 27.88483238220215,
|
| 1080 |
+
"learning_rate": 8.046666666666665e-06,
|
| 1081 |
+
"loss": 0.2335,
|
| 1082 |
+
"step": 3800
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.5640345056403451,
|
| 1086 |
+
"grad_norm": 13.373597145080566,
|
| 1087 |
+
"learning_rate": 7.88e-06,
|
| 1088 |
+
"loss": 0.1962,
|
| 1089 |
+
"step": 3825
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.5677210056772101,
|
| 1093 |
+
"grad_norm": 13.356810569763184,
|
| 1094 |
+
"learning_rate": 7.713333333333335e-06,
|
| 1095 |
+
"loss": 0.2506,
|
| 1096 |
+
"step": 3850
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 0.5714075057140751,
|
| 1100 |
+
"grad_norm": 15.66338062286377,
|
| 1101 |
+
"learning_rate": 7.546666666666666e-06,
|
| 1102 |
+
"loss": 0.186,
|
| 1103 |
+
"step": 3875
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 0.57509400575094,
|
| 1107 |
+
"grad_norm": 8.784326553344727,
|
| 1108 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 1109 |
+
"loss": 0.2063,
|
| 1110 |
+
"step": 3900
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.578780505787805,
|
| 1114 |
+
"grad_norm": 25.057741165161133,
|
| 1115 |
+
"learning_rate": 7.2133333333333334e-06,
|
| 1116 |
+
"loss": 0.1827,
|
| 1117 |
+
"step": 3925
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 0.58246700582467,
|
| 1121 |
+
"grad_norm": 14.403828620910645,
|
| 1122 |
+
"learning_rate": 7.046666666666667e-06,
|
| 1123 |
+
"loss": 0.2408,
|
| 1124 |
+
"step": 3950
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.586153505861535,
|
| 1128 |
+
"grad_norm": 21.203418731689453,
|
| 1129 |
+
"learning_rate": 6.88e-06,
|
| 1130 |
+
"loss": 0.2332,
|
| 1131 |
+
"step": 3975
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.5898400058984,
|
| 1135 |
+
"grad_norm": 9.135237693786621,
|
| 1136 |
+
"learning_rate": 6.713333333333333e-06,
|
| 1137 |
+
"loss": 0.2314,
|
| 1138 |
+
"step": 4000
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.593526505935265,
|
| 1142 |
+
"grad_norm": 19.98188591003418,
|
| 1143 |
+
"learning_rate": 6.546666666666667e-06,
|
| 1144 |
+
"loss": 0.2248,
|
| 1145 |
+
"step": 4025
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.59721300597213,
|
| 1149 |
+
"grad_norm": 19.135929107666016,
|
| 1150 |
+
"learning_rate": 6.38e-06,
|
| 1151 |
+
"loss": 0.2214,
|
| 1152 |
+
"step": 4050
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.6008995060089951,
|
| 1156 |
+
"grad_norm": 15.8812837600708,
|
| 1157 |
+
"learning_rate": 6.213333333333334e-06,
|
| 1158 |
+
"loss": 0.2308,
|
| 1159 |
+
"step": 4075
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 0.6045860060458601,
|
| 1163 |
+
"grad_norm": 10.606598854064941,
|
| 1164 |
+
"learning_rate": 6.046666666666667e-06,
|
| 1165 |
+
"loss": 0.1975,
|
| 1166 |
+
"step": 4100
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.6082725060827251,
|
| 1170 |
+
"grad_norm": 21.68260955810547,
|
| 1171 |
+
"learning_rate": 5.8800000000000005e-06,
|
| 1172 |
+
"loss": 0.2112,
|
| 1173 |
+
"step": 4125
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 0.6119590061195901,
|
| 1177 |
+
"grad_norm": 22.929859161376953,
|
| 1178 |
+
"learning_rate": 5.713333333333333e-06,
|
| 1179 |
+
"loss": 0.2397,
|
| 1180 |
+
"step": 4150
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.6156455061564551,
|
| 1184 |
+
"grad_norm": 14.349271774291992,
|
| 1185 |
+
"learning_rate": 5.546666666666666e-06,
|
| 1186 |
+
"loss": 0.2353,
|
| 1187 |
+
"step": 4175
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 0.6193320061933201,
|
| 1191 |
+
"grad_norm": 16.48160743713379,
|
| 1192 |
+
"learning_rate": 5.38e-06,
|
| 1193 |
+
"loss": 0.2248,
|
| 1194 |
+
"step": 4200
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.6230185062301851,
|
| 1198 |
+
"grad_norm": 22.764707565307617,
|
| 1199 |
+
"learning_rate": 5.213333333333334e-06,
|
| 1200 |
+
"loss": 0.1864,
|
| 1201 |
+
"step": 4225
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.6267050062670501,
|
| 1205 |
+
"grad_norm": 15.090928077697754,
|
| 1206 |
+
"learning_rate": 5.046666666666667e-06,
|
| 1207 |
+
"loss": 0.1842,
|
| 1208 |
+
"step": 4250
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.630391506303915,
|
| 1212 |
+
"grad_norm": 13.214851379394531,
|
| 1213 |
+
"learning_rate": 4.88e-06,
|
| 1214 |
+
"loss": 0.2283,
|
| 1215 |
+
"step": 4275
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 0.63407800634078,
|
| 1219 |
+
"grad_norm": 18.852643966674805,
|
| 1220 |
+
"learning_rate": 4.713333333333334e-06,
|
| 1221 |
+
"loss": 0.1686,
|
| 1222 |
+
"step": 4300
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 0.637764506377645,
|
| 1226 |
+
"grad_norm": 16.26368522644043,
|
| 1227 |
+
"learning_rate": 4.546666666666667e-06,
|
| 1228 |
+
"loss": 0.2027,
|
| 1229 |
+
"step": 4325
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 0.64145100641451,
|
| 1233 |
+
"grad_norm": 13.48696517944336,
|
| 1234 |
+
"learning_rate": 4.3799999999999996e-06,
|
| 1235 |
+
"loss": 0.2064,
|
| 1236 |
+
"step": 4350
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 0.645137506451375,
|
| 1240 |
+
"grad_norm": 20.077945709228516,
|
| 1241 |
+
"learning_rate": 4.213333333333333e-06,
|
| 1242 |
+
"loss": 0.2112,
|
| 1243 |
+
"step": 4375
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 0.64882400648824,
|
| 1247 |
+
"grad_norm": 18.807682037353516,
|
| 1248 |
+
"learning_rate": 4.046666666666667e-06,
|
| 1249 |
+
"loss": 0.2558,
|
| 1250 |
+
"step": 4400
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.6525105065251051,
|
| 1254 |
+
"grad_norm": 22.49479866027832,
|
| 1255 |
+
"learning_rate": 3.88e-06,
|
| 1256 |
+
"loss": 0.2025,
|
| 1257 |
+
"step": 4425
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.6561970065619701,
|
| 1261 |
+
"grad_norm": 18.677318572998047,
|
| 1262 |
+
"learning_rate": 3.7133333333333335e-06,
|
| 1263 |
+
"loss": 0.2006,
|
| 1264 |
+
"step": 4450
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.6598835065988351,
|
| 1268 |
+
"grad_norm": 15.892446517944336,
|
| 1269 |
+
"learning_rate": 3.546666666666667e-06,
|
| 1270 |
+
"loss": 0.1713,
|
| 1271 |
+
"step": 4475
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.6635700066357001,
|
| 1275 |
+
"grad_norm": 10.78048038482666,
|
| 1276 |
+
"learning_rate": 3.38e-06,
|
| 1277 |
+
"loss": 0.1848,
|
| 1278 |
+
"step": 4500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.6672565066725651,
|
| 1282 |
+
"grad_norm": 14.83394718170166,
|
| 1283 |
+
"learning_rate": 3.2133333333333336e-06,
|
| 1284 |
+
"loss": 0.1986,
|
| 1285 |
+
"step": 4525
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.6709430067094301,
|
| 1289 |
+
"grad_norm": 20.701946258544922,
|
| 1290 |
+
"learning_rate": 3.0466666666666666e-06,
|
| 1291 |
+
"loss": 0.259,
|
| 1292 |
+
"step": 4550
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.6746295067462951,
|
| 1296 |
+
"grad_norm": 21.86021614074707,
|
| 1297 |
+
"learning_rate": 2.88e-06,
|
| 1298 |
+
"loss": 0.1949,
|
| 1299 |
+
"step": 4575
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.6783160067831601,
|
| 1303 |
+
"grad_norm": 12.091095924377441,
|
| 1304 |
+
"learning_rate": 2.7133333333333333e-06,
|
| 1305 |
+
"loss": 0.2145,
|
| 1306 |
+
"step": 4600
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.6820025068200251,
|
| 1310 |
+
"grad_norm": 11.082942962646484,
|
| 1311 |
+
"learning_rate": 2.5466666666666667e-06,
|
| 1312 |
+
"loss": 0.2452,
|
| 1313 |
+
"step": 4625
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.68568900685689,
|
| 1317 |
+
"grad_norm": 24.253511428833008,
|
| 1318 |
+
"learning_rate": 2.38e-06,
|
| 1319 |
+
"loss": 0.1859,
|
| 1320 |
+
"step": 4650
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 0.689375506893755,
|
| 1324 |
+
"grad_norm": 15.203668594360352,
|
| 1325 |
+
"learning_rate": 2.2133333333333335e-06,
|
| 1326 |
+
"loss": 0.2019,
|
| 1327 |
+
"step": 4675
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.69306200693062,
|
| 1331 |
+
"grad_norm": 15.922442436218262,
|
| 1332 |
+
"learning_rate": 2.046666666666667e-06,
|
| 1333 |
+
"loss": 0.1888,
|
| 1334 |
+
"step": 4700
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.696748506967485,
|
| 1338 |
+
"grad_norm": 20.253328323364258,
|
| 1339 |
+
"learning_rate": 1.88e-06,
|
| 1340 |
+
"loss": 0.1537,
|
| 1341 |
+
"step": 4725
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.70043500700435,
|
| 1345 |
+
"grad_norm": 31.187671661376953,
|
| 1346 |
+
"learning_rate": 1.7133333333333334e-06,
|
| 1347 |
+
"loss": 0.2385,
|
| 1348 |
+
"step": 4750
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.7041215070412151,
|
| 1352 |
+
"grad_norm": 11.846253395080566,
|
| 1353 |
+
"learning_rate": 1.5466666666666668e-06,
|
| 1354 |
+
"loss": 0.2205,
|
| 1355 |
+
"step": 4775
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.7078080070780801,
|
| 1359 |
+
"grad_norm": 13.351944923400879,
|
| 1360 |
+
"learning_rate": 1.38e-06,
|
| 1361 |
+
"loss": 0.2159,
|
| 1362 |
+
"step": 4800
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 0.7114945071149451,
|
| 1366 |
+
"grad_norm": 15.803413391113281,
|
| 1367 |
+
"learning_rate": 1.2133333333333333e-06,
|
| 1368 |
+
"loss": 0.1816,
|
| 1369 |
+
"step": 4825
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.7151810071518101,
|
| 1373 |
+
"grad_norm": 17.536611557006836,
|
| 1374 |
+
"learning_rate": 1.0466666666666667e-06,
|
| 1375 |
+
"loss": 0.2221,
|
| 1376 |
+
"step": 4850
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.7188675071886751,
|
| 1380 |
+
"grad_norm": 13.475217819213867,
|
| 1381 |
+
"learning_rate": 8.8e-07,
|
| 1382 |
+
"loss": 0.1869,
|
| 1383 |
+
"step": 4875
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 0.7225540072255401,
|
| 1387 |
+
"grad_norm": 19.432355880737305,
|
| 1388 |
+
"learning_rate": 7.133333333333333e-07,
|
| 1389 |
+
"loss": 0.2164,
|
| 1390 |
+
"step": 4900
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.7262405072624051,
|
| 1394 |
+
"grad_norm": 18.54833984375,
|
| 1395 |
+
"learning_rate": 5.466666666666667e-07,
|
| 1396 |
+
"loss": 0.2257,
|
| 1397 |
+
"step": 4925
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 0.7299270072992701,
|
| 1401 |
+
"grad_norm": 17.6931095123291,
|
| 1402 |
+
"learning_rate": 3.8e-07,
|
| 1403 |
+
"loss": 0.1724,
|
| 1404 |
+
"step": 4950
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.7336135073361351,
|
| 1408 |
+
"grad_norm": 8.846336364746094,
|
| 1409 |
+
"learning_rate": 2.1333333333333334e-07,
|
| 1410 |
+
"loss": 0.2273,
|
| 1411 |
+
"step": 4975
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.737300007373,
|
| 1415 |
+
"grad_norm": 11.599397659301758,
|
| 1416 |
+
"learning_rate": 4.666666666666667e-08,
|
| 1417 |
+
"loss": 0.1619,
|
| 1418 |
+
"step": 5000
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.737300007373,
|
| 1422 |
+
"eval_loss": 3.3529720306396484,
|
| 1423 |
+
"eval_runtime": 5174.5819,
|
| 1424 |
+
"eval_samples_per_second": 1.222,
|
| 1425 |
+
"eval_steps_per_second": 0.306,
|
| 1426 |
+
"eval_wer": 1092.1104536489152,
|
| 1427 |
+
"step": 5000
|
| 1428 |
+
}
|
| 1429 |
+
],
|
| 1430 |
+
"logging_steps": 25,
|
| 1431 |
+
"max_steps": 5000,
|
| 1432 |
+
"num_input_tokens_seen": 0,
|
| 1433 |
+
"num_train_epochs": 1,
|
| 1434 |
+
"save_steps": 2500,
|
| 1435 |
+
"stateful_callbacks": {
|
| 1436 |
+
"EarlyStoppingCallback": {
|
| 1437 |
+
"args": {
|
| 1438 |
+
"early_stopping_patience": 2,
|
| 1439 |
+
"early_stopping_threshold": 0.0
|
| 1440 |
+
},
|
| 1441 |
+
"attributes": {
|
| 1442 |
+
"early_stopping_patience_counter": 1
|
| 1443 |
+
}
|
| 1444 |
+
},
|
| 1445 |
+
"TrainerControl": {
|
| 1446 |
+
"args": {
|
| 1447 |
+
"should_epoch_stop": false,
|
| 1448 |
+
"should_evaluate": false,
|
| 1449 |
+
"should_log": false,
|
| 1450 |
+
"should_save": true,
|
| 1451 |
+
"should_training_stop": true
|
| 1452 |
+
},
|
| 1453 |
+
"attributes": {}
|
| 1454 |
+
}
|
| 1455 |
+
},
|
| 1456 |
+
"total_flos": 1.15434160128e+19,
|
| 1457 |
+
"train_batch_size": 4,
|
| 1458 |
+
"trial_name": null,
|
| 1459 |
+
"trial_params": null
|
| 1460 |
+
}
|
checkpoint-5000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49c35fd5ca580f4feb96cbb758e4aefd3fe94f32fc886c38ea6f6f7c699c3ab4
|
| 3 |
+
size 5432
|
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Marwan-Kasem/whisper-small-hi42",
|
| 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": "4.47.0",
|
| 58 |
+
"use_cache": false,
|
| 59 |
+
"use_weighted_layer_sum": false,
|
| 60 |
+
"vocab_size": 51865
|
| 61 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7725fd97010404929ac3190fe96450acde6e0fa5c682bc1d797fd6445141392c
|
| 3 |
+
size 966995080
|
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 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49c35fd5ca580f4feb96cbb758e4aefd3fe94f32fc886c38ea6f6f7c699c3ab4
|
| 3 |
+
size 5432
|