Upload 9 files
Browse files- config.json +52 -0
- generation_config.json +265 -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 +1466 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": [
|
| 11 |
+
220,
|
| 12 |
+
50257
|
| 13 |
+
],
|
| 14 |
+
"bos_token_id": 50257,
|
| 15 |
+
"classifier_proj_size": 256,
|
| 16 |
+
"d_model": 768,
|
| 17 |
+
"decoder_attention_heads": 12,
|
| 18 |
+
"decoder_ffn_dim": 3072,
|
| 19 |
+
"decoder_layerdrop": 0.0,
|
| 20 |
+
"decoder_layers": 12,
|
| 21 |
+
"decoder_start_token_id": 50258,
|
| 22 |
+
"dropout": 0.0,
|
| 23 |
+
"encoder_attention_heads": 12,
|
| 24 |
+
"encoder_ffn_dim": 3072,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 12,
|
| 27 |
+
"eos_token_id": 50257,
|
| 28 |
+
"forced_decoder_ids": null,
|
| 29 |
+
"init_std": 0.02,
|
| 30 |
+
"is_encoder_decoder": true,
|
| 31 |
+
"mask_feature_length": 10,
|
| 32 |
+
"mask_feature_min_masks": 0,
|
| 33 |
+
"mask_feature_prob": 0.0,
|
| 34 |
+
"mask_time_length": 10,
|
| 35 |
+
"mask_time_min_masks": 2,
|
| 36 |
+
"mask_time_prob": 0.05,
|
| 37 |
+
"max_length": 448,
|
| 38 |
+
"max_source_positions": 1500,
|
| 39 |
+
"max_target_positions": 448,
|
| 40 |
+
"median_filter_width": 7,
|
| 41 |
+
"model_type": "whisper",
|
| 42 |
+
"num_hidden_layers": 12,
|
| 43 |
+
"num_mel_bins": 80,
|
| 44 |
+
"pad_token_id": 50257,
|
| 45 |
+
"scale_embedding": false,
|
| 46 |
+
"suppress_tokens": [],
|
| 47 |
+
"torch_dtype": "float32",
|
| 48 |
+
"transformers_version": "4.39.3",
|
| 49 |
+
"use_cache": true,
|
| 50 |
+
"use_weighted_layer_sum": false,
|
| 51 |
+
"vocab_size": 51865
|
| 52 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "ar",
|
| 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 |
+
50358,
|
| 255 |
+
50359,
|
| 256 |
+
50360,
|
| 257 |
+
50361,
|
| 258 |
+
50362
|
| 259 |
+
],
|
| 260 |
+
"task_to_id": {
|
| 261 |
+
"transcribe": 50359,
|
| 262 |
+
"translate": 50358
|
| 263 |
+
},
|
| 264 |
+
"transformers_version": "4.39.3"
|
| 265 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1ae77b72de42f6e09129485431973c1971945f040a6cb8557918f3361cd9625
|
| 3 |
+
size 966995080
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd12cfed7e9489e292cf95fbdee7a50427df6a6d31ce1b1c0edb0f1e77142f15
|
| 3 |
+
size 1925060287
|
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 sha256:65294ba46b43630d4db9e903f777809b34a0613783b89ebcc00f38d1120bb0dc
|
| 3 |
+
size 14503
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c02f28989117220816c4786199a619e1b6d89fce1a536bd36f26f62fdf45377
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1466 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7512307808831327,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-dialect_gulf/checkpoint-5000",
|
| 4 |
+
"epoch": 2.311604253351826,
|
| 5 |
+
"eval_steps": 1000,
|
| 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": 0.01,
|
| 13 |
+
"grad_norm": 90.13593292236328,
|
| 14 |
+
"learning_rate": 5.000000000000001e-07,
|
| 15 |
+
"loss": 4.8868,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 33.31315231323242,
|
| 21 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 22 |
+
"loss": 4.1622,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03,
|
| 27 |
+
"grad_norm": 35.78958511352539,
|
| 28 |
+
"learning_rate": 1.5e-06,
|
| 29 |
+
"loss": 3.3472,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05,
|
| 34 |
+
"grad_norm": 32.040348052978516,
|
| 35 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 36 |
+
"loss": 2.8814,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06,
|
| 41 |
+
"grad_norm": 31.268539428710938,
|
| 42 |
+
"learning_rate": 2.5e-06,
|
| 43 |
+
"loss": 2.4352,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07,
|
| 48 |
+
"grad_norm": 31.179664611816406,
|
| 49 |
+
"learning_rate": 3e-06,
|
| 50 |
+
"loss": 2.3021,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"grad_norm": 23.714216232299805,
|
| 56 |
+
"learning_rate": 3.5e-06,
|
| 57 |
+
"loss": 2.0636,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.09,
|
| 62 |
+
"grad_norm": 27.635934829711914,
|
| 63 |
+
"learning_rate": 4.000000000000001e-06,
|
| 64 |
+
"loss": 1.9784,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1,
|
| 69 |
+
"grad_norm": 23.310068130493164,
|
| 70 |
+
"learning_rate": 4.5e-06,
|
| 71 |
+
"loss": 2.0005,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.12,
|
| 76 |
+
"grad_norm": 22.89619255065918,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 1.9184,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.13,
|
| 83 |
+
"grad_norm": 25.67959976196289,
|
| 84 |
+
"learning_rate": 5.500000000000001e-06,
|
| 85 |
+
"loss": 1.8769,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.14,
|
| 90 |
+
"grad_norm": 24.483964920043945,
|
| 91 |
+
"learning_rate": 6e-06,
|
| 92 |
+
"loss": 1.7286,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.15,
|
| 97 |
+
"grad_norm": 27.483518600463867,
|
| 98 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 99 |
+
"loss": 1.762,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.16,
|
| 104 |
+
"grad_norm": 26.084688186645508,
|
| 105 |
+
"learning_rate": 7e-06,
|
| 106 |
+
"loss": 1.5194,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.17,
|
| 111 |
+
"grad_norm": 23.23436164855957,
|
| 112 |
+
"learning_rate": 7.500000000000001e-06,
|
| 113 |
+
"loss": 1.5613,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.18,
|
| 118 |
+
"grad_norm": 27.090152740478516,
|
| 119 |
+
"learning_rate": 8.000000000000001e-06,
|
| 120 |
+
"loss": 1.3849,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2,
|
| 125 |
+
"grad_norm": 20.59851837158203,
|
| 126 |
+
"learning_rate": 8.5e-06,
|
| 127 |
+
"loss": 1.3407,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.21,
|
| 132 |
+
"grad_norm": 21.621706008911133,
|
| 133 |
+
"learning_rate": 9e-06,
|
| 134 |
+
"loss": 1.4216,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.22,
|
| 139 |
+
"grad_norm": 20.763568878173828,
|
| 140 |
+
"learning_rate": 9.5e-06,
|
| 141 |
+
"loss": 1.3252,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.23,
|
| 146 |
+
"grad_norm": 23.10639762878418,
|
| 147 |
+
"learning_rate": 1e-05,
|
| 148 |
+
"loss": 1.2477,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.24,
|
| 153 |
+
"grad_norm": 21.58993911743164,
|
| 154 |
+
"learning_rate": 9.944444444444445e-06,
|
| 155 |
+
"loss": 1.2779,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.25,
|
| 160 |
+
"grad_norm": 25.219682693481445,
|
| 161 |
+
"learning_rate": 9.88888888888889e-06,
|
| 162 |
+
"loss": 1.4722,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.27,
|
| 167 |
+
"grad_norm": 27.170644760131836,
|
| 168 |
+
"learning_rate": 9.833333333333333e-06,
|
| 169 |
+
"loss": 1.2436,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.28,
|
| 174 |
+
"grad_norm": 21.273935317993164,
|
| 175 |
+
"learning_rate": 9.777777777777779e-06,
|
| 176 |
+
"loss": 1.219,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.29,
|
| 181 |
+
"grad_norm": 28.816940307617188,
|
| 182 |
+
"learning_rate": 9.722222222222223e-06,
|
| 183 |
+
"loss": 1.4722,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3,
|
| 188 |
+
"grad_norm": 28.660903930664062,
|
| 189 |
+
"learning_rate": 9.666666666666667e-06,
|
| 190 |
+
"loss": 1.2638,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.31,
|
| 195 |
+
"grad_norm": 22.813350677490234,
|
| 196 |
+
"learning_rate": 9.611111111111112e-06,
|
| 197 |
+
"loss": 1.4662,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.32,
|
| 202 |
+
"grad_norm": 20.098535537719727,
|
| 203 |
+
"learning_rate": 9.555555555555556e-06,
|
| 204 |
+
"loss": 1.3016,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.34,
|
| 209 |
+
"grad_norm": 23.036521911621094,
|
| 210 |
+
"learning_rate": 9.5e-06,
|
| 211 |
+
"loss": 1.3634,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.35,
|
| 216 |
+
"grad_norm": 27.71617889404297,
|
| 217 |
+
"learning_rate": 9.444444444444445e-06,
|
| 218 |
+
"loss": 1.3038,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.36,
|
| 223 |
+
"grad_norm": 22.73453140258789,
|
| 224 |
+
"learning_rate": 9.38888888888889e-06,
|
| 225 |
+
"loss": 1.2634,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.37,
|
| 230 |
+
"grad_norm": 26.822202682495117,
|
| 231 |
+
"learning_rate": 9.333333333333334e-06,
|
| 232 |
+
"loss": 1.28,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.38,
|
| 237 |
+
"grad_norm": 23.140613555908203,
|
| 238 |
+
"learning_rate": 9.277777777777778e-06,
|
| 239 |
+
"loss": 1.3511,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.39,
|
| 244 |
+
"grad_norm": 25.16455078125,
|
| 245 |
+
"learning_rate": 9.222222222222224e-06,
|
| 246 |
+
"loss": 1.2653,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.4,
|
| 251 |
+
"grad_norm": 26.37599754333496,
|
| 252 |
+
"learning_rate": 9.166666666666666e-06,
|
| 253 |
+
"loss": 1.3031,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.42,
|
| 258 |
+
"grad_norm": 24.899930953979492,
|
| 259 |
+
"learning_rate": 9.111111111111112e-06,
|
| 260 |
+
"loss": 1.2706,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.43,
|
| 265 |
+
"grad_norm": 22.268238067626953,
|
| 266 |
+
"learning_rate": 9.055555555555556e-06,
|
| 267 |
+
"loss": 1.3006,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.44,
|
| 272 |
+
"grad_norm": 20.132190704345703,
|
| 273 |
+
"learning_rate": 9e-06,
|
| 274 |
+
"loss": 1.2915,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.45,
|
| 279 |
+
"grad_norm": 21.949169158935547,
|
| 280 |
+
"learning_rate": 8.944444444444446e-06,
|
| 281 |
+
"loss": 1.2265,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.46,
|
| 286 |
+
"grad_norm": 26.4661922454834,
|
| 287 |
+
"learning_rate": 8.888888888888888e-06,
|
| 288 |
+
"loss": 1.3905,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.46,
|
| 293 |
+
"eval_loss": 1.2462494373321533,
|
| 294 |
+
"eval_runtime": 1237.1982,
|
| 295 |
+
"eval_samples_per_second": 3.497,
|
| 296 |
+
"eval_steps_per_second": 0.437,
|
| 297 |
+
"eval_wer": 0.903961220934636,
|
| 298 |
+
"step": 1000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.47,
|
| 302 |
+
"grad_norm": 20.229887008666992,
|
| 303 |
+
"learning_rate": 8.833333333333334e-06,
|
| 304 |
+
"loss": 1.2907,
|
| 305 |
+
"step": 1025
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.49,
|
| 309 |
+
"grad_norm": 22.81121253967285,
|
| 310 |
+
"learning_rate": 8.777777777777778e-06,
|
| 311 |
+
"loss": 1.3117,
|
| 312 |
+
"step": 1050
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.5,
|
| 316 |
+
"grad_norm": 23.232364654541016,
|
| 317 |
+
"learning_rate": 8.722222222222224e-06,
|
| 318 |
+
"loss": 1.2393,
|
| 319 |
+
"step": 1075
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.51,
|
| 323 |
+
"grad_norm": 19.813980102539062,
|
| 324 |
+
"learning_rate": 8.666666666666668e-06,
|
| 325 |
+
"loss": 1.253,
|
| 326 |
+
"step": 1100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.52,
|
| 330 |
+
"grad_norm": 22.089860916137695,
|
| 331 |
+
"learning_rate": 8.611111111111112e-06,
|
| 332 |
+
"loss": 1.2862,
|
| 333 |
+
"step": 1125
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.53,
|
| 337 |
+
"grad_norm": 24.085248947143555,
|
| 338 |
+
"learning_rate": 8.555555555555556e-06,
|
| 339 |
+
"loss": 1.3403,
|
| 340 |
+
"step": 1150
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.54,
|
| 344 |
+
"grad_norm": 21.81845474243164,
|
| 345 |
+
"learning_rate": 8.5e-06,
|
| 346 |
+
"loss": 1.2885,
|
| 347 |
+
"step": 1175
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.55,
|
| 351 |
+
"grad_norm": 23.052608489990234,
|
| 352 |
+
"learning_rate": 8.444444444444446e-06,
|
| 353 |
+
"loss": 1.3301,
|
| 354 |
+
"step": 1200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.57,
|
| 358 |
+
"grad_norm": 18.781269073486328,
|
| 359 |
+
"learning_rate": 8.38888888888889e-06,
|
| 360 |
+
"loss": 1.3634,
|
| 361 |
+
"step": 1225
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.58,
|
| 365 |
+
"grad_norm": 19.929943084716797,
|
| 366 |
+
"learning_rate": 8.333333333333334e-06,
|
| 367 |
+
"loss": 1.2226,
|
| 368 |
+
"step": 1250
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.59,
|
| 372 |
+
"grad_norm": 30.602449417114258,
|
| 373 |
+
"learning_rate": 8.277777777777778e-06,
|
| 374 |
+
"loss": 1.3299,
|
| 375 |
+
"step": 1275
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.6,
|
| 379 |
+
"grad_norm": 23.87131118774414,
|
| 380 |
+
"learning_rate": 8.222222222222222e-06,
|
| 381 |
+
"loss": 1.2976,
|
| 382 |
+
"step": 1300
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.61,
|
| 386 |
+
"grad_norm": 22.464336395263672,
|
| 387 |
+
"learning_rate": 8.166666666666668e-06,
|
| 388 |
+
"loss": 1.2414,
|
| 389 |
+
"step": 1325
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.62,
|
| 393 |
+
"grad_norm": 17.31785774230957,
|
| 394 |
+
"learning_rate": 8.111111111111112e-06,
|
| 395 |
+
"loss": 1.221,
|
| 396 |
+
"step": 1350
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.64,
|
| 400 |
+
"grad_norm": 26.171546936035156,
|
| 401 |
+
"learning_rate": 8.055555555555557e-06,
|
| 402 |
+
"loss": 1.2388,
|
| 403 |
+
"step": 1375
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.65,
|
| 407 |
+
"grad_norm": 19.773033142089844,
|
| 408 |
+
"learning_rate": 8.000000000000001e-06,
|
| 409 |
+
"loss": 1.1601,
|
| 410 |
+
"step": 1400
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.66,
|
| 414 |
+
"grad_norm": 31.665481567382812,
|
| 415 |
+
"learning_rate": 7.944444444444445e-06,
|
| 416 |
+
"loss": 1.3741,
|
| 417 |
+
"step": 1425
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.67,
|
| 421 |
+
"grad_norm": 24.492671966552734,
|
| 422 |
+
"learning_rate": 7.88888888888889e-06,
|
| 423 |
+
"loss": 1.2745,
|
| 424 |
+
"step": 1450
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.68,
|
| 428 |
+
"grad_norm": 24.3986873626709,
|
| 429 |
+
"learning_rate": 7.833333333333333e-06,
|
| 430 |
+
"loss": 1.2371,
|
| 431 |
+
"step": 1475
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.69,
|
| 435 |
+
"grad_norm": 22.881649017333984,
|
| 436 |
+
"learning_rate": 7.77777777777778e-06,
|
| 437 |
+
"loss": 1.254,
|
| 438 |
+
"step": 1500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.71,
|
| 442 |
+
"grad_norm": 23.861434936523438,
|
| 443 |
+
"learning_rate": 7.722222222222223e-06,
|
| 444 |
+
"loss": 1.1979,
|
| 445 |
+
"step": 1525
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.72,
|
| 449 |
+
"grad_norm": 20.938926696777344,
|
| 450 |
+
"learning_rate": 7.666666666666667e-06,
|
| 451 |
+
"loss": 1.1873,
|
| 452 |
+
"step": 1550
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.73,
|
| 456 |
+
"grad_norm": 31.408279418945312,
|
| 457 |
+
"learning_rate": 7.611111111111111e-06,
|
| 458 |
+
"loss": 1.1919,
|
| 459 |
+
"step": 1575
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.74,
|
| 463 |
+
"grad_norm": 23.88611602783203,
|
| 464 |
+
"learning_rate": 7.555555555555556e-06,
|
| 465 |
+
"loss": 1.2749,
|
| 466 |
+
"step": 1600
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.75,
|
| 470 |
+
"grad_norm": 20.98951530456543,
|
| 471 |
+
"learning_rate": 7.500000000000001e-06,
|
| 472 |
+
"loss": 1.0798,
|
| 473 |
+
"step": 1625
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.76,
|
| 477 |
+
"grad_norm": 19.691837310791016,
|
| 478 |
+
"learning_rate": 7.444444444444445e-06,
|
| 479 |
+
"loss": 1.1863,
|
| 480 |
+
"step": 1650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.77,
|
| 484 |
+
"grad_norm": 18.230724334716797,
|
| 485 |
+
"learning_rate": 7.38888888888889e-06,
|
| 486 |
+
"loss": 1.2122,
|
| 487 |
+
"step": 1675
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.79,
|
| 491 |
+
"grad_norm": 18.965030670166016,
|
| 492 |
+
"learning_rate": 7.333333333333333e-06,
|
| 493 |
+
"loss": 1.147,
|
| 494 |
+
"step": 1700
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.8,
|
| 498 |
+
"grad_norm": 26.649892807006836,
|
| 499 |
+
"learning_rate": 7.277777777777778e-06,
|
| 500 |
+
"loss": 1.3043,
|
| 501 |
+
"step": 1725
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.81,
|
| 505 |
+
"grad_norm": 19.97573471069336,
|
| 506 |
+
"learning_rate": 7.222222222222223e-06,
|
| 507 |
+
"loss": 1.18,
|
| 508 |
+
"step": 1750
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.82,
|
| 512 |
+
"grad_norm": 18.884883880615234,
|
| 513 |
+
"learning_rate": 7.166666666666667e-06,
|
| 514 |
+
"loss": 1.1741,
|
| 515 |
+
"step": 1775
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.83,
|
| 519 |
+
"grad_norm": 21.117107391357422,
|
| 520 |
+
"learning_rate": 7.111111111111112e-06,
|
| 521 |
+
"loss": 1.1664,
|
| 522 |
+
"step": 1800
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.84,
|
| 526 |
+
"grad_norm": 28.765390396118164,
|
| 527 |
+
"learning_rate": 7.055555555555557e-06,
|
| 528 |
+
"loss": 1.3056,
|
| 529 |
+
"step": 1825
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.86,
|
| 533 |
+
"grad_norm": 22.94930076599121,
|
| 534 |
+
"learning_rate": 7e-06,
|
| 535 |
+
"loss": 1.153,
|
| 536 |
+
"step": 1850
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.87,
|
| 540 |
+
"grad_norm": 22.228673934936523,
|
| 541 |
+
"learning_rate": 6.944444444444445e-06,
|
| 542 |
+
"loss": 1.1835,
|
| 543 |
+
"step": 1875
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.88,
|
| 547 |
+
"grad_norm": 27.348892211914062,
|
| 548 |
+
"learning_rate": 6.88888888888889e-06,
|
| 549 |
+
"loss": 1.2132,
|
| 550 |
+
"step": 1900
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.89,
|
| 554 |
+
"grad_norm": 19.449748992919922,
|
| 555 |
+
"learning_rate": 6.833333333333334e-06,
|
| 556 |
+
"loss": 1.2001,
|
| 557 |
+
"step": 1925
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.9,
|
| 561 |
+
"grad_norm": 23.61025047302246,
|
| 562 |
+
"learning_rate": 6.777777777777779e-06,
|
| 563 |
+
"loss": 1.0774,
|
| 564 |
+
"step": 1950
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.91,
|
| 568 |
+
"grad_norm": 24.239519119262695,
|
| 569 |
+
"learning_rate": 6.7222222222222235e-06,
|
| 570 |
+
"loss": 1.1935,
|
| 571 |
+
"step": 1975
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.92,
|
| 575 |
+
"grad_norm": 19.700180053710938,
|
| 576 |
+
"learning_rate": 6.666666666666667e-06,
|
| 577 |
+
"loss": 1.3338,
|
| 578 |
+
"step": 2000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.92,
|
| 582 |
+
"eval_loss": 1.1501092910766602,
|
| 583 |
+
"eval_runtime": 1305.57,
|
| 584 |
+
"eval_samples_per_second": 3.313,
|
| 585 |
+
"eval_steps_per_second": 0.414,
|
| 586 |
+
"eval_wer": 0.902900855866091,
|
| 587 |
+
"step": 2000
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 0.94,
|
| 591 |
+
"grad_norm": 24.58094024658203,
|
| 592 |
+
"learning_rate": 6.6111111111111115e-06,
|
| 593 |
+
"loss": 1.1615,
|
| 594 |
+
"step": 2025
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.95,
|
| 598 |
+
"grad_norm": 30.994125366210938,
|
| 599 |
+
"learning_rate": 6.555555555555556e-06,
|
| 600 |
+
"loss": 1.2499,
|
| 601 |
+
"step": 2050
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.96,
|
| 605 |
+
"grad_norm": 22.892292022705078,
|
| 606 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 607 |
+
"loss": 1.1866,
|
| 608 |
+
"step": 2075
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.97,
|
| 612 |
+
"grad_norm": 23.727901458740234,
|
| 613 |
+
"learning_rate": 6.444444444444445e-06,
|
| 614 |
+
"loss": 1.241,
|
| 615 |
+
"step": 2100
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.98,
|
| 619 |
+
"grad_norm": 22.032814025878906,
|
| 620 |
+
"learning_rate": 6.3888888888888885e-06,
|
| 621 |
+
"loss": 1.2042,
|
| 622 |
+
"step": 2125
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.99,
|
| 626 |
+
"grad_norm": 29.404563903808594,
|
| 627 |
+
"learning_rate": 6.333333333333333e-06,
|
| 628 |
+
"loss": 1.1603,
|
| 629 |
+
"step": 2150
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 1.01,
|
| 633 |
+
"grad_norm": 14.76396656036377,
|
| 634 |
+
"learning_rate": 6.277777777777778e-06,
|
| 635 |
+
"loss": 1.0143,
|
| 636 |
+
"step": 2175
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 1.02,
|
| 640 |
+
"grad_norm": 13.789732933044434,
|
| 641 |
+
"learning_rate": 6.222222222222223e-06,
|
| 642 |
+
"loss": 0.8707,
|
| 643 |
+
"step": 2200
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 1.03,
|
| 647 |
+
"grad_norm": 21.95280647277832,
|
| 648 |
+
"learning_rate": 6.166666666666667e-06,
|
| 649 |
+
"loss": 0.9127,
|
| 650 |
+
"step": 2225
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 1.04,
|
| 654 |
+
"grad_norm": 16.230443954467773,
|
| 655 |
+
"learning_rate": 6.111111111111112e-06,
|
| 656 |
+
"loss": 0.8598,
|
| 657 |
+
"step": 2250
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 1.05,
|
| 661 |
+
"grad_norm": 19.354671478271484,
|
| 662 |
+
"learning_rate": 6.055555555555555e-06,
|
| 663 |
+
"loss": 0.8387,
|
| 664 |
+
"step": 2275
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 1.06,
|
| 668 |
+
"grad_norm": 18.098474502563477,
|
| 669 |
+
"learning_rate": 6e-06,
|
| 670 |
+
"loss": 0.7913,
|
| 671 |
+
"step": 2300
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 1.07,
|
| 675 |
+
"grad_norm": 16.24185562133789,
|
| 676 |
+
"learning_rate": 5.944444444444445e-06,
|
| 677 |
+
"loss": 0.8831,
|
| 678 |
+
"step": 2325
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 1.09,
|
| 682 |
+
"grad_norm": 15.916351318359375,
|
| 683 |
+
"learning_rate": 5.88888888888889e-06,
|
| 684 |
+
"loss": 0.9592,
|
| 685 |
+
"step": 2350
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 1.1,
|
| 689 |
+
"grad_norm": 19.874614715576172,
|
| 690 |
+
"learning_rate": 5.833333333333334e-06,
|
| 691 |
+
"loss": 0.869,
|
| 692 |
+
"step": 2375
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.11,
|
| 696 |
+
"grad_norm": 14.805360794067383,
|
| 697 |
+
"learning_rate": 5.777777777777778e-06,
|
| 698 |
+
"loss": 0.7478,
|
| 699 |
+
"step": 2400
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.12,
|
| 703 |
+
"grad_norm": 24.80560874938965,
|
| 704 |
+
"learning_rate": 5.722222222222222e-06,
|
| 705 |
+
"loss": 0.9983,
|
| 706 |
+
"step": 2425
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 1.13,
|
| 710 |
+
"grad_norm": 14.018525123596191,
|
| 711 |
+
"learning_rate": 5.666666666666667e-06,
|
| 712 |
+
"loss": 0.9365,
|
| 713 |
+
"step": 2450
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.14,
|
| 717 |
+
"grad_norm": 16.925121307373047,
|
| 718 |
+
"learning_rate": 5.611111111111112e-06,
|
| 719 |
+
"loss": 0.8463,
|
| 720 |
+
"step": 2475
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.16,
|
| 724 |
+
"grad_norm": 12.009042739868164,
|
| 725 |
+
"learning_rate": 5.555555555555557e-06,
|
| 726 |
+
"loss": 0.7888,
|
| 727 |
+
"step": 2500
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.17,
|
| 731 |
+
"grad_norm": 17.96883773803711,
|
| 732 |
+
"learning_rate": 5.500000000000001e-06,
|
| 733 |
+
"loss": 1.0053,
|
| 734 |
+
"step": 2525
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 1.18,
|
| 738 |
+
"grad_norm": 26.6944522857666,
|
| 739 |
+
"learning_rate": 5.444444444444445e-06,
|
| 740 |
+
"loss": 0.9585,
|
| 741 |
+
"step": 2550
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 1.19,
|
| 745 |
+
"grad_norm": 17.153728485107422,
|
| 746 |
+
"learning_rate": 5.388888888888889e-06,
|
| 747 |
+
"loss": 0.9195,
|
| 748 |
+
"step": 2575
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 1.2,
|
| 752 |
+
"grad_norm": 18.12151527404785,
|
| 753 |
+
"learning_rate": 5.333333333333334e-06,
|
| 754 |
+
"loss": 0.8074,
|
| 755 |
+
"step": 2600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 1.21,
|
| 759 |
+
"grad_norm": 15.750487327575684,
|
| 760 |
+
"learning_rate": 5.2777777777777785e-06,
|
| 761 |
+
"loss": 0.9239,
|
| 762 |
+
"step": 2625
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.23,
|
| 766 |
+
"grad_norm": 17.816286087036133,
|
| 767 |
+
"learning_rate": 5.2222222222222226e-06,
|
| 768 |
+
"loss": 0.9279,
|
| 769 |
+
"step": 2650
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 1.24,
|
| 773 |
+
"grad_norm": 21.26312255859375,
|
| 774 |
+
"learning_rate": 5.1666666666666675e-06,
|
| 775 |
+
"loss": 0.8332,
|
| 776 |
+
"step": 2675
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 1.25,
|
| 780 |
+
"grad_norm": 13.23405647277832,
|
| 781 |
+
"learning_rate": 5.1111111111111115e-06,
|
| 782 |
+
"loss": 0.8514,
|
| 783 |
+
"step": 2700
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 1.26,
|
| 787 |
+
"grad_norm": 14.837157249450684,
|
| 788 |
+
"learning_rate": 5.0555555555555555e-06,
|
| 789 |
+
"loss": 0.8468,
|
| 790 |
+
"step": 2725
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 1.27,
|
| 794 |
+
"grad_norm": 16.646862030029297,
|
| 795 |
+
"learning_rate": 5e-06,
|
| 796 |
+
"loss": 0.8662,
|
| 797 |
+
"step": 2750
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.28,
|
| 801 |
+
"grad_norm": 18.599933624267578,
|
| 802 |
+
"learning_rate": 4.944444444444445e-06,
|
| 803 |
+
"loss": 1.0057,
|
| 804 |
+
"step": 2775
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.29,
|
| 808 |
+
"grad_norm": 18.52560043334961,
|
| 809 |
+
"learning_rate": 4.888888888888889e-06,
|
| 810 |
+
"loss": 0.8529,
|
| 811 |
+
"step": 2800
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 1.31,
|
| 815 |
+
"grad_norm": 23.720380783081055,
|
| 816 |
+
"learning_rate": 4.833333333333333e-06,
|
| 817 |
+
"loss": 0.9172,
|
| 818 |
+
"step": 2825
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 1.32,
|
| 822 |
+
"grad_norm": 16.967018127441406,
|
| 823 |
+
"learning_rate": 4.777777777777778e-06,
|
| 824 |
+
"loss": 0.8931,
|
| 825 |
+
"step": 2850
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 1.33,
|
| 829 |
+
"grad_norm": 20.544633865356445,
|
| 830 |
+
"learning_rate": 4.722222222222222e-06,
|
| 831 |
+
"loss": 0.8951,
|
| 832 |
+
"step": 2875
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 1.34,
|
| 836 |
+
"grad_norm": 24.084787368774414,
|
| 837 |
+
"learning_rate": 4.666666666666667e-06,
|
| 838 |
+
"loss": 0.8265,
|
| 839 |
+
"step": 2900
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 1.35,
|
| 843 |
+
"grad_norm": 23.94223976135254,
|
| 844 |
+
"learning_rate": 4.611111111111112e-06,
|
| 845 |
+
"loss": 0.9139,
|
| 846 |
+
"step": 2925
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 1.36,
|
| 850 |
+
"grad_norm": 19.657983779907227,
|
| 851 |
+
"learning_rate": 4.555555555555556e-06,
|
| 852 |
+
"loss": 0.8641,
|
| 853 |
+
"step": 2950
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.38,
|
| 857 |
+
"grad_norm": 21.78879737854004,
|
| 858 |
+
"learning_rate": 4.5e-06,
|
| 859 |
+
"loss": 0.9701,
|
| 860 |
+
"step": 2975
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.39,
|
| 864 |
+
"grad_norm": 15.626680374145508,
|
| 865 |
+
"learning_rate": 4.444444444444444e-06,
|
| 866 |
+
"loss": 0.9277,
|
| 867 |
+
"step": 3000
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 1.39,
|
| 871 |
+
"eval_loss": 1.1189059019088745,
|
| 872 |
+
"eval_runtime": 1297.5862,
|
| 873 |
+
"eval_samples_per_second": 3.334,
|
| 874 |
+
"eval_steps_per_second": 0.417,
|
| 875 |
+
"eval_wer": 0.818488222373703,
|
| 876 |
+
"step": 3000
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.4,
|
| 880 |
+
"grad_norm": 16.269102096557617,
|
| 881 |
+
"learning_rate": 4.388888888888889e-06,
|
| 882 |
+
"loss": 0.8269,
|
| 883 |
+
"step": 3025
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.41,
|
| 887 |
+
"grad_norm": 16.133930206298828,
|
| 888 |
+
"learning_rate": 4.333333333333334e-06,
|
| 889 |
+
"loss": 0.8246,
|
| 890 |
+
"step": 3050
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.42,
|
| 894 |
+
"grad_norm": 23.846628189086914,
|
| 895 |
+
"learning_rate": 4.277777777777778e-06,
|
| 896 |
+
"loss": 0.9221,
|
| 897 |
+
"step": 3075
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.43,
|
| 901 |
+
"grad_norm": 14.437520027160645,
|
| 902 |
+
"learning_rate": 4.222222222222223e-06,
|
| 903 |
+
"loss": 0.9663,
|
| 904 |
+
"step": 3100
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.44,
|
| 908 |
+
"grad_norm": 15.177382469177246,
|
| 909 |
+
"learning_rate": 4.166666666666667e-06,
|
| 910 |
+
"loss": 0.8925,
|
| 911 |
+
"step": 3125
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.46,
|
| 915 |
+
"grad_norm": 18.592479705810547,
|
| 916 |
+
"learning_rate": 4.111111111111111e-06,
|
| 917 |
+
"loss": 0.8471,
|
| 918 |
+
"step": 3150
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 1.47,
|
| 922 |
+
"grad_norm": 16.767436981201172,
|
| 923 |
+
"learning_rate": 4.055555555555556e-06,
|
| 924 |
+
"loss": 0.8792,
|
| 925 |
+
"step": 3175
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 1.48,
|
| 929 |
+
"grad_norm": 25.8040828704834,
|
| 930 |
+
"learning_rate": 4.000000000000001e-06,
|
| 931 |
+
"loss": 0.9201,
|
| 932 |
+
"step": 3200
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.49,
|
| 936 |
+
"grad_norm": 21.563386917114258,
|
| 937 |
+
"learning_rate": 3.944444444444445e-06,
|
| 938 |
+
"loss": 0.833,
|
| 939 |
+
"step": 3225
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.5,
|
| 943 |
+
"grad_norm": 18.131933212280273,
|
| 944 |
+
"learning_rate": 3.88888888888889e-06,
|
| 945 |
+
"loss": 0.8592,
|
| 946 |
+
"step": 3250
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.51,
|
| 950 |
+
"grad_norm": 19.301654815673828,
|
| 951 |
+
"learning_rate": 3.833333333333334e-06,
|
| 952 |
+
"loss": 0.9349,
|
| 953 |
+
"step": 3275
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 1.53,
|
| 957 |
+
"grad_norm": 21.040206909179688,
|
| 958 |
+
"learning_rate": 3.777777777777778e-06,
|
| 959 |
+
"loss": 0.8683,
|
| 960 |
+
"step": 3300
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.54,
|
| 964 |
+
"grad_norm": 17.75210952758789,
|
| 965 |
+
"learning_rate": 3.7222222222222225e-06,
|
| 966 |
+
"loss": 0.8643,
|
| 967 |
+
"step": 3325
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 1.55,
|
| 971 |
+
"grad_norm": 15.362386703491211,
|
| 972 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 973 |
+
"loss": 0.8959,
|
| 974 |
+
"step": 3350
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.56,
|
| 978 |
+
"grad_norm": 17.963134765625,
|
| 979 |
+
"learning_rate": 3.6111111111111115e-06,
|
| 980 |
+
"loss": 0.9468,
|
| 981 |
+
"step": 3375
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 1.57,
|
| 985 |
+
"grad_norm": 22.818912506103516,
|
| 986 |
+
"learning_rate": 3.555555555555556e-06,
|
| 987 |
+
"loss": 0.8257,
|
| 988 |
+
"step": 3400
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.58,
|
| 992 |
+
"grad_norm": 17.68602752685547,
|
| 993 |
+
"learning_rate": 3.5e-06,
|
| 994 |
+
"loss": 0.8878,
|
| 995 |
+
"step": 3425
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 1.6,
|
| 999 |
+
"grad_norm": 15.19880199432373,
|
| 1000 |
+
"learning_rate": 3.444444444444445e-06,
|
| 1001 |
+
"loss": 0.8924,
|
| 1002 |
+
"step": 3450
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 1.61,
|
| 1006 |
+
"grad_norm": 21.873252868652344,
|
| 1007 |
+
"learning_rate": 3.3888888888888893e-06,
|
| 1008 |
+
"loss": 0.8991,
|
| 1009 |
+
"step": 3475
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.62,
|
| 1013 |
+
"grad_norm": 18.36728286743164,
|
| 1014 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 1015 |
+
"loss": 0.8458,
|
| 1016 |
+
"step": 3500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.63,
|
| 1020 |
+
"grad_norm": 18.431533813476562,
|
| 1021 |
+
"learning_rate": 3.277777777777778e-06,
|
| 1022 |
+
"loss": 0.7883,
|
| 1023 |
+
"step": 3525
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 1.64,
|
| 1027 |
+
"grad_norm": 13.323966026306152,
|
| 1028 |
+
"learning_rate": 3.2222222222222227e-06,
|
| 1029 |
+
"loss": 0.808,
|
| 1030 |
+
"step": 3550
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.65,
|
| 1034 |
+
"grad_norm": 20.740156173706055,
|
| 1035 |
+
"learning_rate": 3.1666666666666667e-06,
|
| 1036 |
+
"loss": 0.8501,
|
| 1037 |
+
"step": 3575
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 1.66,
|
| 1041 |
+
"grad_norm": 17.27450180053711,
|
| 1042 |
+
"learning_rate": 3.1111111111111116e-06,
|
| 1043 |
+
"loss": 0.862,
|
| 1044 |
+
"step": 3600
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.68,
|
| 1048 |
+
"grad_norm": 23.401592254638672,
|
| 1049 |
+
"learning_rate": 3.055555555555556e-06,
|
| 1050 |
+
"loss": 0.8988,
|
| 1051 |
+
"step": 3625
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.69,
|
| 1055 |
+
"grad_norm": 18.699295043945312,
|
| 1056 |
+
"learning_rate": 3e-06,
|
| 1057 |
+
"loss": 0.8809,
|
| 1058 |
+
"step": 3650
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.7,
|
| 1062 |
+
"grad_norm": 17.551536560058594,
|
| 1063 |
+
"learning_rate": 2.944444444444445e-06,
|
| 1064 |
+
"loss": 0.8908,
|
| 1065 |
+
"step": 3675
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.71,
|
| 1069 |
+
"grad_norm": 16.441694259643555,
|
| 1070 |
+
"learning_rate": 2.888888888888889e-06,
|
| 1071 |
+
"loss": 0.875,
|
| 1072 |
+
"step": 3700
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.72,
|
| 1076 |
+
"grad_norm": 27.084482192993164,
|
| 1077 |
+
"learning_rate": 2.8333333333333335e-06,
|
| 1078 |
+
"loss": 0.9332,
|
| 1079 |
+
"step": 3725
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.73,
|
| 1083 |
+
"grad_norm": 15.829773902893066,
|
| 1084 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 1085 |
+
"loss": 0.8631,
|
| 1086 |
+
"step": 3750
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 1.75,
|
| 1090 |
+
"grad_norm": 17.43021583557129,
|
| 1091 |
+
"learning_rate": 2.7222222222222224e-06,
|
| 1092 |
+
"loss": 0.835,
|
| 1093 |
+
"step": 3775
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 1.76,
|
| 1097 |
+
"grad_norm": 16.65917205810547,
|
| 1098 |
+
"learning_rate": 2.666666666666667e-06,
|
| 1099 |
+
"loss": 0.8623,
|
| 1100 |
+
"step": 3800
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.77,
|
| 1104 |
+
"grad_norm": 20.004430770874023,
|
| 1105 |
+
"learning_rate": 2.6111111111111113e-06,
|
| 1106 |
+
"loss": 0.89,
|
| 1107 |
+
"step": 3825
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 1.78,
|
| 1111 |
+
"grad_norm": 19.59697151184082,
|
| 1112 |
+
"learning_rate": 2.5555555555555557e-06,
|
| 1113 |
+
"loss": 0.9355,
|
| 1114 |
+
"step": 3850
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.79,
|
| 1118 |
+
"grad_norm": 15.853771209716797,
|
| 1119 |
+
"learning_rate": 2.5e-06,
|
| 1120 |
+
"loss": 0.8388,
|
| 1121 |
+
"step": 3875
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.8,
|
| 1125 |
+
"grad_norm": 18.28623390197754,
|
| 1126 |
+
"learning_rate": 2.4444444444444447e-06,
|
| 1127 |
+
"loss": 0.866,
|
| 1128 |
+
"step": 3900
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.81,
|
| 1132 |
+
"grad_norm": 12.830389976501465,
|
| 1133 |
+
"learning_rate": 2.388888888888889e-06,
|
| 1134 |
+
"loss": 0.7384,
|
| 1135 |
+
"step": 3925
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 1.83,
|
| 1139 |
+
"grad_norm": 20.37066078186035,
|
| 1140 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 1141 |
+
"loss": 0.7555,
|
| 1142 |
+
"step": 3950
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.84,
|
| 1146 |
+
"grad_norm": 15.657346725463867,
|
| 1147 |
+
"learning_rate": 2.277777777777778e-06,
|
| 1148 |
+
"loss": 0.8481,
|
| 1149 |
+
"step": 3975
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 1.85,
|
| 1153 |
+
"grad_norm": 17.99932098388672,
|
| 1154 |
+
"learning_rate": 2.222222222222222e-06,
|
| 1155 |
+
"loss": 0.8347,
|
| 1156 |
+
"step": 4000
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 1.85,
|
| 1160 |
+
"eval_loss": 1.0929533243179321,
|
| 1161 |
+
"eval_runtime": 1274.136,
|
| 1162 |
+
"eval_samples_per_second": 3.395,
|
| 1163 |
+
"eval_steps_per_second": 0.425,
|
| 1164 |
+
"eval_wer": 0.8011058092857684,
|
| 1165 |
+
"step": 4000
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.86,
|
| 1169 |
+
"grad_norm": 18.715171813964844,
|
| 1170 |
+
"learning_rate": 2.166666666666667e-06,
|
| 1171 |
+
"loss": 0.9483,
|
| 1172 |
+
"step": 4025
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.87,
|
| 1176 |
+
"grad_norm": 14.89651870727539,
|
| 1177 |
+
"learning_rate": 2.1111111111111114e-06,
|
| 1178 |
+
"loss": 0.829,
|
| 1179 |
+
"step": 4050
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.88,
|
| 1183 |
+
"grad_norm": 13.295916557312012,
|
| 1184 |
+
"learning_rate": 2.0555555555555555e-06,
|
| 1185 |
+
"loss": 0.7805,
|
| 1186 |
+
"step": 4075
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.9,
|
| 1190 |
+
"grad_norm": 19.08588218688965,
|
| 1191 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 1192 |
+
"loss": 0.7778,
|
| 1193 |
+
"step": 4100
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.91,
|
| 1197 |
+
"grad_norm": 12.506399154663086,
|
| 1198 |
+
"learning_rate": 1.944444444444445e-06,
|
| 1199 |
+
"loss": 0.7863,
|
| 1200 |
+
"step": 4125
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.92,
|
| 1204 |
+
"grad_norm": 18.24089241027832,
|
| 1205 |
+
"learning_rate": 1.888888888888889e-06,
|
| 1206 |
+
"loss": 0.7969,
|
| 1207 |
+
"step": 4150
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.93,
|
| 1211 |
+
"grad_norm": 20.995338439941406,
|
| 1212 |
+
"learning_rate": 1.8333333333333333e-06,
|
| 1213 |
+
"loss": 0.9124,
|
| 1214 |
+
"step": 4175
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.94,
|
| 1218 |
+
"grad_norm": 16.891035079956055,
|
| 1219 |
+
"learning_rate": 1.777777777777778e-06,
|
| 1220 |
+
"loss": 0.8769,
|
| 1221 |
+
"step": 4200
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.95,
|
| 1225 |
+
"grad_norm": 18.63488006591797,
|
| 1226 |
+
"learning_rate": 1.7222222222222224e-06,
|
| 1227 |
+
"loss": 0.8952,
|
| 1228 |
+
"step": 4225
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.96,
|
| 1232 |
+
"grad_norm": 19.848390579223633,
|
| 1233 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 1234 |
+
"loss": 0.8361,
|
| 1235 |
+
"step": 4250
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.98,
|
| 1239 |
+
"grad_norm": 18.764118194580078,
|
| 1240 |
+
"learning_rate": 1.6111111111111113e-06,
|
| 1241 |
+
"loss": 0.795,
|
| 1242 |
+
"step": 4275
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.99,
|
| 1246 |
+
"grad_norm": 20.843151092529297,
|
| 1247 |
+
"learning_rate": 1.5555555555555558e-06,
|
| 1248 |
+
"loss": 0.9876,
|
| 1249 |
+
"step": 4300
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 2.0,
|
| 1253 |
+
"grad_norm": 19.12489891052246,
|
| 1254 |
+
"learning_rate": 1.5e-06,
|
| 1255 |
+
"loss": 0.8053,
|
| 1256 |
+
"step": 4325
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.01,
|
| 1260 |
+
"grad_norm": 14.411953926086426,
|
| 1261 |
+
"learning_rate": 1.4444444444444445e-06,
|
| 1262 |
+
"loss": 0.5857,
|
| 1263 |
+
"step": 4350
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.02,
|
| 1267 |
+
"grad_norm": 21.73296546936035,
|
| 1268 |
+
"learning_rate": 1.3888888888888892e-06,
|
| 1269 |
+
"loss": 0.7082,
|
| 1270 |
+
"step": 4375
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.03,
|
| 1274 |
+
"grad_norm": 18.120756149291992,
|
| 1275 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 1276 |
+
"loss": 0.6726,
|
| 1277 |
+
"step": 4400
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.05,
|
| 1281 |
+
"grad_norm": 12.758247375488281,
|
| 1282 |
+
"learning_rate": 1.2777777777777779e-06,
|
| 1283 |
+
"loss": 0.6329,
|
| 1284 |
+
"step": 4425
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.06,
|
| 1288 |
+
"grad_norm": 10.172853469848633,
|
| 1289 |
+
"learning_rate": 1.2222222222222223e-06,
|
| 1290 |
+
"loss": 0.5615,
|
| 1291 |
+
"step": 4450
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.07,
|
| 1295 |
+
"grad_norm": 18.662038803100586,
|
| 1296 |
+
"learning_rate": 1.1666666666666668e-06,
|
| 1297 |
+
"loss": 0.5792,
|
| 1298 |
+
"step": 4475
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 2.08,
|
| 1302 |
+
"grad_norm": 12.439136505126953,
|
| 1303 |
+
"learning_rate": 1.111111111111111e-06,
|
| 1304 |
+
"loss": 0.7267,
|
| 1305 |
+
"step": 4500
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 2.09,
|
| 1309 |
+
"grad_norm": 16.754854202270508,
|
| 1310 |
+
"learning_rate": 1.0555555555555557e-06,
|
| 1311 |
+
"loss": 0.6498,
|
| 1312 |
+
"step": 4525
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 2.1,
|
| 1316 |
+
"grad_norm": 16.47597312927246,
|
| 1317 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 1318 |
+
"loss": 0.6845,
|
| 1319 |
+
"step": 4550
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 2.12,
|
| 1323 |
+
"grad_norm": 16.60858154296875,
|
| 1324 |
+
"learning_rate": 9.444444444444445e-07,
|
| 1325 |
+
"loss": 0.5911,
|
| 1326 |
+
"step": 4575
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 2.13,
|
| 1330 |
+
"grad_norm": 12.026427268981934,
|
| 1331 |
+
"learning_rate": 8.88888888888889e-07,
|
| 1332 |
+
"loss": 0.5721,
|
| 1333 |
+
"step": 4600
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 2.14,
|
| 1337 |
+
"grad_norm": 18.53736686706543,
|
| 1338 |
+
"learning_rate": 8.333333333333333e-07,
|
| 1339 |
+
"loss": 0.66,
|
| 1340 |
+
"step": 4625
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.15,
|
| 1344 |
+
"grad_norm": 16.173511505126953,
|
| 1345 |
+
"learning_rate": 7.777777777777779e-07,
|
| 1346 |
+
"loss": 0.6813,
|
| 1347 |
+
"step": 4650
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 2.16,
|
| 1351 |
+
"grad_norm": 14.248098373413086,
|
| 1352 |
+
"learning_rate": 7.222222222222222e-07,
|
| 1353 |
+
"loss": 0.6902,
|
| 1354 |
+
"step": 4675
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 2.17,
|
| 1358 |
+
"grad_norm": 18.32253646850586,
|
| 1359 |
+
"learning_rate": 6.666666666666667e-07,
|
| 1360 |
+
"loss": 0.6995,
|
| 1361 |
+
"step": 4700
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 2.18,
|
| 1365 |
+
"grad_norm": 10.451536178588867,
|
| 1366 |
+
"learning_rate": 6.111111111111112e-07,
|
| 1367 |
+
"loss": 0.6376,
|
| 1368 |
+
"step": 4725
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 2.2,
|
| 1372 |
+
"grad_norm": 8.597134590148926,
|
| 1373 |
+
"learning_rate": 5.555555555555555e-07,
|
| 1374 |
+
"loss": 0.6471,
|
| 1375 |
+
"step": 4750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 2.21,
|
| 1379 |
+
"grad_norm": 12.938069343566895,
|
| 1380 |
+
"learning_rate": 5.000000000000001e-07,
|
| 1381 |
+
"loss": 0.658,
|
| 1382 |
+
"step": 4775
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.22,
|
| 1386 |
+
"grad_norm": 12.209549903869629,
|
| 1387 |
+
"learning_rate": 4.444444444444445e-07,
|
| 1388 |
+
"loss": 0.636,
|
| 1389 |
+
"step": 4800
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.23,
|
| 1393 |
+
"grad_norm": 9.981619834899902,
|
| 1394 |
+
"learning_rate": 3.8888888888888895e-07,
|
| 1395 |
+
"loss": 0.5447,
|
| 1396 |
+
"step": 4825
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.24,
|
| 1400 |
+
"grad_norm": 17.577558517456055,
|
| 1401 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 1402 |
+
"loss": 0.6757,
|
| 1403 |
+
"step": 4850
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 2.25,
|
| 1407 |
+
"grad_norm": 17.589126586914062,
|
| 1408 |
+
"learning_rate": 2.7777777777777776e-07,
|
| 1409 |
+
"loss": 0.6394,
|
| 1410 |
+
"step": 4875
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 2.27,
|
| 1414 |
+
"grad_norm": 18.11832618713379,
|
| 1415 |
+
"learning_rate": 2.2222222222222224e-07,
|
| 1416 |
+
"loss": 0.6869,
|
| 1417 |
+
"step": 4900
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 2.28,
|
| 1421 |
+
"grad_norm": 14.544137954711914,
|
| 1422 |
+
"learning_rate": 1.6666666666666668e-07,
|
| 1423 |
+
"loss": 0.6432,
|
| 1424 |
+
"step": 4925
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.29,
|
| 1428 |
+
"grad_norm": 15.83905029296875,
|
| 1429 |
+
"learning_rate": 1.1111111111111112e-07,
|
| 1430 |
+
"loss": 0.609,
|
| 1431 |
+
"step": 4950
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 2.3,
|
| 1435 |
+
"grad_norm": 19.050613403320312,
|
| 1436 |
+
"learning_rate": 5.555555555555556e-08,
|
| 1437 |
+
"loss": 0.6702,
|
| 1438 |
+
"step": 4975
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.31,
|
| 1442 |
+
"grad_norm": 17.937570571899414,
|
| 1443 |
+
"learning_rate": 0.0,
|
| 1444 |
+
"loss": 0.6948,
|
| 1445 |
+
"step": 5000
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.31,
|
| 1449 |
+
"eval_loss": 1.1064683198928833,
|
| 1450 |
+
"eval_runtime": 1237.1908,
|
| 1451 |
+
"eval_samples_per_second": 3.497,
|
| 1452 |
+
"eval_steps_per_second": 0.437,
|
| 1453 |
+
"eval_wer": 0.7512307808831327,
|
| 1454 |
+
"step": 5000
|
| 1455 |
+
}
|
| 1456 |
+
],
|
| 1457 |
+
"logging_steps": 25,
|
| 1458 |
+
"max_steps": 5000,
|
| 1459 |
+
"num_input_tokens_seen": 0,
|
| 1460 |
+
"num_train_epochs": 3,
|
| 1461 |
+
"save_steps": 1000,
|
| 1462 |
+
"total_flos": 1.154283884199936e+19,
|
| 1463 |
+
"train_batch_size": 8,
|
| 1464 |
+
"trial_name": null,
|
| 1465 |
+
"trial_params": null
|
| 1466 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8a8053647fc94dd5b2d3c4b97f2518ff0ace819857f3b8ea95e7a7b3c977f34
|
| 3 |
+
size 4591
|