Training in progress, step 5000, checkpoint
Browse files- checkpoint-5000/config.json +65 -0
- checkpoint-5000/generation_config.json +265 -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 +1511 -0
- checkpoint-5000/training_args.bin +3 -0
checkpoint-5000/config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/kaggle/input/whisper-small-final/whisper-small-final",
|
| 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": [
|
| 29 |
+
[
|
| 30 |
+
1,
|
| 31 |
+
50272
|
| 32 |
+
],
|
| 33 |
+
[
|
| 34 |
+
2,
|
| 35 |
+
50359
|
| 36 |
+
],
|
| 37 |
+
[
|
| 38 |
+
3,
|
| 39 |
+
50363
|
| 40 |
+
]
|
| 41 |
+
],
|
| 42 |
+
"init_std": 0.02,
|
| 43 |
+
"is_encoder_decoder": true,
|
| 44 |
+
"mask_feature_length": 10,
|
| 45 |
+
"mask_feature_min_masks": 0,
|
| 46 |
+
"mask_feature_prob": 0.0,
|
| 47 |
+
"mask_time_length": 10,
|
| 48 |
+
"mask_time_min_masks": 2,
|
| 49 |
+
"mask_time_prob": 0.05,
|
| 50 |
+
"max_length": 448,
|
| 51 |
+
"max_source_positions": 1500,
|
| 52 |
+
"max_target_positions": 448,
|
| 53 |
+
"median_filter_width": 7,
|
| 54 |
+
"model_type": "whisper",
|
| 55 |
+
"num_hidden_layers": 12,
|
| 56 |
+
"num_mel_bins": 80,
|
| 57 |
+
"pad_token_id": 50257,
|
| 58 |
+
"scale_embedding": false,
|
| 59 |
+
"suppress_tokens": [],
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.40.0.dev0",
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"use_weighted_layer_sum": false,
|
| 64 |
+
"vocab_size": 51865
|
| 65 |
+
}
|
checkpoint-5000/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": "4.40.0.dev0"
|
| 265 |
+
}
|
checkpoint-5000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4eb0a5e09ac00248f0deab40230108fc0d0d66866dc288224d87d40f7177b00
|
| 3 |
+
size 966995080
|
checkpoint-5000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b4b45b7e307d9dc41998287fed30d379cb79316a1cca09f97b138f2e1c3120fb
|
| 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:e843b7e254516bc2d8a50f20455e32b6fd8226072dcf8d97d475be384506beae
|
| 3 |
+
size 14244
|
checkpoint-5000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c1ae66fa59eb3053c17fc09748378de509c08eceaa840c4fc94d0b3854402b57
|
| 3 |
+
size 1064
|
checkpoint-5000/trainer_state.json
ADDED
|
@@ -0,0 +1,1511 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.9224686264397457,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-final-v2/checkpoint-5000",
|
| 4 |
+
"epoch": 16.181229773462782,
|
| 5 |
+
"eval_steps": 500,
|
| 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": 0.08,
|
| 13 |
+
"grad_norm": 0.0023542253766208887,
|
| 14 |
+
"learning_rate": 5.000000000000001e-07,
|
| 15 |
+
"loss": 0.0001,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.16,
|
| 20 |
+
"grad_norm": 0.003053858410567045,
|
| 21 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 22 |
+
"loss": 0.0001,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.24,
|
| 27 |
+
"grad_norm": 0.007697729859501123,
|
| 28 |
+
"learning_rate": 1.5e-06,
|
| 29 |
+
"loss": 0.0001,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.32,
|
| 34 |
+
"grad_norm": 0.01041096169501543,
|
| 35 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 36 |
+
"loss": 0.0,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.4,
|
| 41 |
+
"grad_norm": 0.0023387186229228973,
|
| 42 |
+
"learning_rate": 2.5e-06,
|
| 43 |
+
"loss": 0.0,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.49,
|
| 48 |
+
"grad_norm": 0.0036297033075243235,
|
| 49 |
+
"learning_rate": 3e-06,
|
| 50 |
+
"loss": 0.0001,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.57,
|
| 55 |
+
"grad_norm": 0.0016004816861823201,
|
| 56 |
+
"learning_rate": 3.5e-06,
|
| 57 |
+
"loss": 0.0,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.65,
|
| 62 |
+
"grad_norm": 0.0035209895577281713,
|
| 63 |
+
"learning_rate": 4.000000000000001e-06,
|
| 64 |
+
"loss": 0.0,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.73,
|
| 69 |
+
"grad_norm": 0.37100252509117126,
|
| 70 |
+
"learning_rate": 4.5e-06,
|
| 71 |
+
"loss": 0.0015,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.81,
|
| 76 |
+
"grad_norm": 0.2593393325805664,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.003,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.89,
|
| 83 |
+
"grad_norm": 1.1273244619369507,
|
| 84 |
+
"learning_rate": 5.500000000000001e-06,
|
| 85 |
+
"loss": 0.003,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.97,
|
| 90 |
+
"grad_norm": 0.11678028106689453,
|
| 91 |
+
"learning_rate": 6e-06,
|
| 92 |
+
"loss": 0.0014,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.05,
|
| 97 |
+
"grad_norm": 2.602248191833496,
|
| 98 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 99 |
+
"loss": 0.0025,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.13,
|
| 104 |
+
"grad_norm": 1.053270697593689,
|
| 105 |
+
"learning_rate": 7e-06,
|
| 106 |
+
"loss": 0.0021,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.21,
|
| 111 |
+
"grad_norm": 0.25527653098106384,
|
| 112 |
+
"learning_rate": 7.500000000000001e-06,
|
| 113 |
+
"loss": 0.0037,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.29,
|
| 118 |
+
"grad_norm": 0.208439901471138,
|
| 119 |
+
"learning_rate": 8.000000000000001e-06,
|
| 120 |
+
"loss": 0.002,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.38,
|
| 125 |
+
"grad_norm": 0.14851850271224976,
|
| 126 |
+
"learning_rate": 8.5e-06,
|
| 127 |
+
"loss": 0.0034,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.46,
|
| 132 |
+
"grad_norm": 1.6750054359436035,
|
| 133 |
+
"learning_rate": 9e-06,
|
| 134 |
+
"loss": 0.0046,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.54,
|
| 139 |
+
"grad_norm": 4.503875255584717,
|
| 140 |
+
"learning_rate": 9.48e-06,
|
| 141 |
+
"loss": 0.0093,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.62,
|
| 146 |
+
"grad_norm": 0.6791089177131653,
|
| 147 |
+
"learning_rate": 9.980000000000001e-06,
|
| 148 |
+
"loss": 0.0073,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.62,
|
| 153 |
+
"eval_loss": 0.024947550147771835,
|
| 154 |
+
"eval_runtime": 794.9124,
|
| 155 |
+
"eval_samples_per_second": 1.551,
|
| 156 |
+
"eval_steps_per_second": 0.195,
|
| 157 |
+
"eval_wer": 5.002578648788035,
|
| 158 |
+
"step": 500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.7,
|
| 162 |
+
"grad_norm": 0.7141709923744202,
|
| 163 |
+
"learning_rate": 9.931428571428571e-06,
|
| 164 |
+
"loss": 0.0022,
|
| 165 |
+
"step": 525
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.78,
|
| 169 |
+
"grad_norm": 1.1480615139007568,
|
| 170 |
+
"learning_rate": 9.86e-06,
|
| 171 |
+
"loss": 0.0036,
|
| 172 |
+
"step": 550
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.86,
|
| 176 |
+
"grad_norm": 1.9711138010025024,
|
| 177 |
+
"learning_rate": 9.78857142857143e-06,
|
| 178 |
+
"loss": 0.0043,
|
| 179 |
+
"step": 575
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.94,
|
| 183 |
+
"grad_norm": 1.0525217056274414,
|
| 184 |
+
"learning_rate": 9.717142857142858e-06,
|
| 185 |
+
"loss": 0.0051,
|
| 186 |
+
"step": 600
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.02,
|
| 190 |
+
"grad_norm": 0.15985222160816193,
|
| 191 |
+
"learning_rate": 9.645714285714286e-06,
|
| 192 |
+
"loss": 0.0057,
|
| 193 |
+
"step": 625
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.1,
|
| 197 |
+
"grad_norm": 0.8261871933937073,
|
| 198 |
+
"learning_rate": 9.574285714285715e-06,
|
| 199 |
+
"loss": 0.0023,
|
| 200 |
+
"step": 650
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.18,
|
| 204 |
+
"grad_norm": 0.04620926454663277,
|
| 205 |
+
"learning_rate": 9.502857142857144e-06,
|
| 206 |
+
"loss": 0.0019,
|
| 207 |
+
"step": 675
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.27,
|
| 211 |
+
"grad_norm": 0.3023536503314972,
|
| 212 |
+
"learning_rate": 9.431428571428573e-06,
|
| 213 |
+
"loss": 0.0012,
|
| 214 |
+
"step": 700
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.35,
|
| 218 |
+
"grad_norm": 0.11112522333860397,
|
| 219 |
+
"learning_rate": 9.360000000000002e-06,
|
| 220 |
+
"loss": 0.0024,
|
| 221 |
+
"step": 725
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.43,
|
| 225 |
+
"grad_norm": 1.6971714496612549,
|
| 226 |
+
"learning_rate": 9.28857142857143e-06,
|
| 227 |
+
"loss": 0.0031,
|
| 228 |
+
"step": 750
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.51,
|
| 232 |
+
"grad_norm": 0.22176721692085266,
|
| 233 |
+
"learning_rate": 9.217142857142858e-06,
|
| 234 |
+
"loss": 0.0026,
|
| 235 |
+
"step": 775
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.59,
|
| 239 |
+
"grad_norm": 0.2839481830596924,
|
| 240 |
+
"learning_rate": 9.145714285714287e-06,
|
| 241 |
+
"loss": 0.0016,
|
| 242 |
+
"step": 800
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.67,
|
| 246 |
+
"grad_norm": 0.6732001900672913,
|
| 247 |
+
"learning_rate": 9.074285714285716e-06,
|
| 248 |
+
"loss": 0.0044,
|
| 249 |
+
"step": 825
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.75,
|
| 253 |
+
"grad_norm": 2.15289044380188,
|
| 254 |
+
"learning_rate": 9.002857142857144e-06,
|
| 255 |
+
"loss": 0.0028,
|
| 256 |
+
"step": 850
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.83,
|
| 260 |
+
"grad_norm": 0.4701692759990692,
|
| 261 |
+
"learning_rate": 8.931428571428573e-06,
|
| 262 |
+
"loss": 0.0009,
|
| 263 |
+
"step": 875
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.91,
|
| 267 |
+
"grad_norm": 0.38268008828163147,
|
| 268 |
+
"learning_rate": 8.860000000000002e-06,
|
| 269 |
+
"loss": 0.0012,
|
| 270 |
+
"step": 900
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.99,
|
| 274 |
+
"grad_norm": 0.057314854115247726,
|
| 275 |
+
"learning_rate": 8.788571428571429e-06,
|
| 276 |
+
"loss": 0.001,
|
| 277 |
+
"step": 925
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 3.07,
|
| 281 |
+
"grad_norm": 0.048205479979515076,
|
| 282 |
+
"learning_rate": 8.717142857142858e-06,
|
| 283 |
+
"loss": 0.0022,
|
| 284 |
+
"step": 950
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.16,
|
| 288 |
+
"grad_norm": 0.008282452821731567,
|
| 289 |
+
"learning_rate": 8.645714285714287e-06,
|
| 290 |
+
"loss": 0.0006,
|
| 291 |
+
"step": 975
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 3.24,
|
| 295 |
+
"grad_norm": 1.3108314275741577,
|
| 296 |
+
"learning_rate": 8.574285714285714e-06,
|
| 297 |
+
"loss": 0.0014,
|
| 298 |
+
"step": 1000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 3.24,
|
| 302 |
+
"eval_loss": 0.02143564447760582,
|
| 303 |
+
"eval_runtime": 802.0669,
|
| 304 |
+
"eval_samples_per_second": 1.537,
|
| 305 |
+
"eval_steps_per_second": 0.193,
|
| 306 |
+
"eval_wer": 4.108647068935878,
|
| 307 |
+
"step": 1000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.32,
|
| 311 |
+
"grad_norm": 0.888531506061554,
|
| 312 |
+
"learning_rate": 8.502857142857143e-06,
|
| 313 |
+
"loss": 0.0025,
|
| 314 |
+
"step": 1025
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.4,
|
| 318 |
+
"grad_norm": 0.013467811048030853,
|
| 319 |
+
"learning_rate": 8.431428571428572e-06,
|
| 320 |
+
"loss": 0.0009,
|
| 321 |
+
"step": 1050
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.48,
|
| 325 |
+
"grad_norm": 1.272592306137085,
|
| 326 |
+
"learning_rate": 8.36e-06,
|
| 327 |
+
"loss": 0.001,
|
| 328 |
+
"step": 1075
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.56,
|
| 332 |
+
"grad_norm": 0.022731788456439972,
|
| 333 |
+
"learning_rate": 8.288571428571429e-06,
|
| 334 |
+
"loss": 0.0002,
|
| 335 |
+
"step": 1100
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.64,
|
| 339 |
+
"grad_norm": 0.49200737476348877,
|
| 340 |
+
"learning_rate": 8.217142857142858e-06,
|
| 341 |
+
"loss": 0.0011,
|
| 342 |
+
"step": 1125
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 3.72,
|
| 346 |
+
"grad_norm": 0.011444230563938618,
|
| 347 |
+
"learning_rate": 8.145714285714287e-06,
|
| 348 |
+
"loss": 0.0005,
|
| 349 |
+
"step": 1150
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 3.8,
|
| 353 |
+
"grad_norm": 0.02571061998605728,
|
| 354 |
+
"learning_rate": 8.074285714285714e-06,
|
| 355 |
+
"loss": 0.0002,
|
| 356 |
+
"step": 1175
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 3.88,
|
| 360 |
+
"grad_norm": 0.02670286037027836,
|
| 361 |
+
"learning_rate": 8.002857142857143e-06,
|
| 362 |
+
"loss": 0.0005,
|
| 363 |
+
"step": 1200
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.96,
|
| 367 |
+
"grad_norm": 0.04544826224446297,
|
| 368 |
+
"learning_rate": 7.931428571428572e-06,
|
| 369 |
+
"loss": 0.0007,
|
| 370 |
+
"step": 1225
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 4.05,
|
| 374 |
+
"grad_norm": 0.028520189225673676,
|
| 375 |
+
"learning_rate": 7.860000000000001e-06,
|
| 376 |
+
"loss": 0.0003,
|
| 377 |
+
"step": 1250
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.13,
|
| 381 |
+
"grad_norm": 0.017675643786787987,
|
| 382 |
+
"learning_rate": 7.788571428571428e-06,
|
| 383 |
+
"loss": 0.0006,
|
| 384 |
+
"step": 1275
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 4.21,
|
| 388 |
+
"grad_norm": 0.16999584436416626,
|
| 389 |
+
"learning_rate": 7.717142857142857e-06,
|
| 390 |
+
"loss": 0.0005,
|
| 391 |
+
"step": 1300
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 4.29,
|
| 395 |
+
"grad_norm": 0.22153036296367645,
|
| 396 |
+
"learning_rate": 7.645714285714286e-06,
|
| 397 |
+
"loss": 0.0004,
|
| 398 |
+
"step": 1325
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.37,
|
| 402 |
+
"grad_norm": 0.0049127149395644665,
|
| 403 |
+
"learning_rate": 7.574285714285715e-06,
|
| 404 |
+
"loss": 0.0001,
|
| 405 |
+
"step": 1350
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.45,
|
| 409 |
+
"grad_norm": 0.02039247751235962,
|
| 410 |
+
"learning_rate": 7.502857142857144e-06,
|
| 411 |
+
"loss": 0.0003,
|
| 412 |
+
"step": 1375
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 4.53,
|
| 416 |
+
"grad_norm": 0.009562624618411064,
|
| 417 |
+
"learning_rate": 7.431428571428572e-06,
|
| 418 |
+
"loss": 0.0002,
|
| 419 |
+
"step": 1400
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.61,
|
| 423 |
+
"grad_norm": 0.019309034571051598,
|
| 424 |
+
"learning_rate": 7.360000000000001e-06,
|
| 425 |
+
"loss": 0.0001,
|
| 426 |
+
"step": 1425
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 4.69,
|
| 430 |
+
"grad_norm": 0.004860865883529186,
|
| 431 |
+
"learning_rate": 7.28857142857143e-06,
|
| 432 |
+
"loss": 0.0008,
|
| 433 |
+
"step": 1450
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 4.77,
|
| 437 |
+
"grad_norm": 0.005680860485881567,
|
| 438 |
+
"learning_rate": 7.217142857142858e-06,
|
| 439 |
+
"loss": 0.0004,
|
| 440 |
+
"step": 1475
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 4.85,
|
| 444 |
+
"grad_norm": 0.02605484426021576,
|
| 445 |
+
"learning_rate": 7.145714285714286e-06,
|
| 446 |
+
"loss": 0.0008,
|
| 447 |
+
"step": 1500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.85,
|
| 451 |
+
"eval_loss": 0.022091694176197052,
|
| 452 |
+
"eval_runtime": 791.3534,
|
| 453 |
+
"eval_samples_per_second": 1.558,
|
| 454 |
+
"eval_steps_per_second": 0.196,
|
| 455 |
+
"eval_wer": 3.988310125494241,
|
| 456 |
+
"step": 1500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.94,
|
| 460 |
+
"grad_norm": 1.7420192956924438,
|
| 461 |
+
"learning_rate": 7.074285714285715e-06,
|
| 462 |
+
"loss": 0.0005,
|
| 463 |
+
"step": 1525
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 5.02,
|
| 467 |
+
"grad_norm": 0.15281762182712555,
|
| 468 |
+
"learning_rate": 7.002857142857143e-06,
|
| 469 |
+
"loss": 0.0004,
|
| 470 |
+
"step": 1550
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 5.1,
|
| 474 |
+
"grad_norm": 0.013328706845641136,
|
| 475 |
+
"learning_rate": 6.931428571428572e-06,
|
| 476 |
+
"loss": 0.0002,
|
| 477 |
+
"step": 1575
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 5.18,
|
| 481 |
+
"grad_norm": 0.23917707800865173,
|
| 482 |
+
"learning_rate": 6.860000000000001e-06,
|
| 483 |
+
"loss": 0.0019,
|
| 484 |
+
"step": 1600
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 5.26,
|
| 488 |
+
"grad_norm": 0.02083628997206688,
|
| 489 |
+
"learning_rate": 6.7885714285714286e-06,
|
| 490 |
+
"loss": 0.0003,
|
| 491 |
+
"step": 1625
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.34,
|
| 495 |
+
"grad_norm": 0.7623525857925415,
|
| 496 |
+
"learning_rate": 6.7171428571428576e-06,
|
| 497 |
+
"loss": 0.0006,
|
| 498 |
+
"step": 1650
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 5.42,
|
| 502 |
+
"grad_norm": 1.761312484741211,
|
| 503 |
+
"learning_rate": 6.645714285714287e-06,
|
| 504 |
+
"loss": 0.0005,
|
| 505 |
+
"step": 1675
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.5,
|
| 509 |
+
"grad_norm": 0.02908872812986374,
|
| 510 |
+
"learning_rate": 6.574285714285716e-06,
|
| 511 |
+
"loss": 0.0003,
|
| 512 |
+
"step": 1700
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.58,
|
| 516 |
+
"grad_norm": 0.0062005845829844475,
|
| 517 |
+
"learning_rate": 6.502857142857143e-06,
|
| 518 |
+
"loss": 0.0003,
|
| 519 |
+
"step": 1725
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.66,
|
| 523 |
+
"grad_norm": 0.009553766809403896,
|
| 524 |
+
"learning_rate": 6.431428571428572e-06,
|
| 525 |
+
"loss": 0.0001,
|
| 526 |
+
"step": 1750
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.74,
|
| 530 |
+
"grad_norm": 0.006461838725954294,
|
| 531 |
+
"learning_rate": 6.360000000000001e-06,
|
| 532 |
+
"loss": 0.0001,
|
| 533 |
+
"step": 1775
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 5.83,
|
| 537 |
+
"grad_norm": 0.006093865260481834,
|
| 538 |
+
"learning_rate": 6.288571428571429e-06,
|
| 539 |
+
"loss": 0.0003,
|
| 540 |
+
"step": 1800
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.91,
|
| 544 |
+
"grad_norm": 0.006265025120228529,
|
| 545 |
+
"learning_rate": 6.217142857142857e-06,
|
| 546 |
+
"loss": 0.0005,
|
| 547 |
+
"step": 1825
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 5.99,
|
| 551 |
+
"grad_norm": 0.08442416042089462,
|
| 552 |
+
"learning_rate": 6.145714285714286e-06,
|
| 553 |
+
"loss": 0.0002,
|
| 554 |
+
"step": 1850
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 6.07,
|
| 558 |
+
"grad_norm": 0.005628472194075584,
|
| 559 |
+
"learning_rate": 6.0742857142857145e-06,
|
| 560 |
+
"loss": 0.0001,
|
| 561 |
+
"step": 1875
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 6.15,
|
| 565 |
+
"grad_norm": 0.0012941138120368123,
|
| 566 |
+
"learning_rate": 6.0028571428571435e-06,
|
| 567 |
+
"loss": 0.0001,
|
| 568 |
+
"step": 1900
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 6.23,
|
| 572 |
+
"grad_norm": 0.003825143212452531,
|
| 573 |
+
"learning_rate": 5.9314285714285725e-06,
|
| 574 |
+
"loss": 0.0001,
|
| 575 |
+
"step": 1925
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 6.31,
|
| 579 |
+
"grad_norm": 0.002939299214631319,
|
| 580 |
+
"learning_rate": 5.86e-06,
|
| 581 |
+
"loss": 0.0,
|
| 582 |
+
"step": 1950
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 6.39,
|
| 586 |
+
"grad_norm": 0.0030452311038970947,
|
| 587 |
+
"learning_rate": 5.788571428571429e-06,
|
| 588 |
+
"loss": 0.0001,
|
| 589 |
+
"step": 1975
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 6.47,
|
| 593 |
+
"grad_norm": 0.017602646723389626,
|
| 594 |
+
"learning_rate": 5.717142857142858e-06,
|
| 595 |
+
"loss": 0.0,
|
| 596 |
+
"step": 2000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 6.47,
|
| 600 |
+
"eval_loss": 0.018006248399615288,
|
| 601 |
+
"eval_runtime": 791.4436,
|
| 602 |
+
"eval_samples_per_second": 1.558,
|
| 603 |
+
"eval_steps_per_second": 0.196,
|
| 604 |
+
"eval_wer": 2.974041602200447,
|
| 605 |
+
"step": 2000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 6.55,
|
| 609 |
+
"grad_norm": 0.09079013019800186,
|
| 610 |
+
"learning_rate": 5.645714285714287e-06,
|
| 611 |
+
"loss": 0.0001,
|
| 612 |
+
"step": 2025
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 6.63,
|
| 616 |
+
"grad_norm": 0.011517921462655067,
|
| 617 |
+
"learning_rate": 5.574285714285714e-06,
|
| 618 |
+
"loss": 0.0001,
|
| 619 |
+
"step": 2050
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 6.72,
|
| 623 |
+
"grad_norm": 0.0036222347989678383,
|
| 624 |
+
"learning_rate": 5.502857142857143e-06,
|
| 625 |
+
"loss": 0.0,
|
| 626 |
+
"step": 2075
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 6.8,
|
| 630 |
+
"grad_norm": 0.09321127831935883,
|
| 631 |
+
"learning_rate": 5.431428571428572e-06,
|
| 632 |
+
"loss": 0.0,
|
| 633 |
+
"step": 2100
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.88,
|
| 637 |
+
"grad_norm": 0.019542288035154343,
|
| 638 |
+
"learning_rate": 5.36e-06,
|
| 639 |
+
"loss": 0.0,
|
| 640 |
+
"step": 2125
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 6.96,
|
| 644 |
+
"grad_norm": 0.010700986720621586,
|
| 645 |
+
"learning_rate": 5.2885714285714285e-06,
|
| 646 |
+
"loss": 0.0001,
|
| 647 |
+
"step": 2150
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 7.04,
|
| 651 |
+
"grad_norm": 0.005260708276182413,
|
| 652 |
+
"learning_rate": 5.2171428571428575e-06,
|
| 653 |
+
"loss": 0.0,
|
| 654 |
+
"step": 2175
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 7.12,
|
| 658 |
+
"grad_norm": 0.0018403552239760756,
|
| 659 |
+
"learning_rate": 5.145714285714286e-06,
|
| 660 |
+
"loss": 0.0,
|
| 661 |
+
"step": 2200
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 7.2,
|
| 665 |
+
"grad_norm": 0.00291392276994884,
|
| 666 |
+
"learning_rate": 5.074285714285715e-06,
|
| 667 |
+
"loss": 0.0,
|
| 668 |
+
"step": 2225
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 7.28,
|
| 672 |
+
"grad_norm": 0.001351356622762978,
|
| 673 |
+
"learning_rate": 5.002857142857144e-06,
|
| 674 |
+
"loss": 0.0,
|
| 675 |
+
"step": 2250
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 7.36,
|
| 679 |
+
"grad_norm": 0.002212280174717307,
|
| 680 |
+
"learning_rate": 4.931428571428572e-06,
|
| 681 |
+
"loss": 0.0,
|
| 682 |
+
"step": 2275
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 7.44,
|
| 686 |
+
"grad_norm": 0.0013704759767279029,
|
| 687 |
+
"learning_rate": 4.86e-06,
|
| 688 |
+
"loss": 0.0,
|
| 689 |
+
"step": 2300
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 7.52,
|
| 693 |
+
"grad_norm": 0.004209516104310751,
|
| 694 |
+
"learning_rate": 4.788571428571429e-06,
|
| 695 |
+
"loss": 0.0,
|
| 696 |
+
"step": 2325
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 7.61,
|
| 700 |
+
"grad_norm": 0.004072306212037802,
|
| 701 |
+
"learning_rate": 4.717142857142857e-06,
|
| 702 |
+
"loss": 0.0,
|
| 703 |
+
"step": 2350
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 7.69,
|
| 707 |
+
"grad_norm": 0.0258013978600502,
|
| 708 |
+
"learning_rate": 4.645714285714286e-06,
|
| 709 |
+
"loss": 0.0,
|
| 710 |
+
"step": 2375
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 7.77,
|
| 714 |
+
"grad_norm": 0.0018390474142506719,
|
| 715 |
+
"learning_rate": 4.574285714285714e-06,
|
| 716 |
+
"loss": 0.0,
|
| 717 |
+
"step": 2400
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 7.85,
|
| 721 |
+
"grad_norm": 0.002616502810269594,
|
| 722 |
+
"learning_rate": 4.5028571428571434e-06,
|
| 723 |
+
"loss": 0.0,
|
| 724 |
+
"step": 2425
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 7.93,
|
| 728 |
+
"grad_norm": 0.0016982157249003649,
|
| 729 |
+
"learning_rate": 4.431428571428572e-06,
|
| 730 |
+
"loss": 0.0,
|
| 731 |
+
"step": 2450
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 8.01,
|
| 735 |
+
"grad_norm": 0.0023189985658973455,
|
| 736 |
+
"learning_rate": 4.360000000000001e-06,
|
| 737 |
+
"loss": 0.0,
|
| 738 |
+
"step": 2475
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 8.09,
|
| 742 |
+
"grad_norm": 0.003353666979819536,
|
| 743 |
+
"learning_rate": 4.288571428571429e-06,
|
| 744 |
+
"loss": 0.0,
|
| 745 |
+
"step": 2500
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 8.09,
|
| 749 |
+
"eval_loss": 0.017685044556856155,
|
| 750 |
+
"eval_runtime": 798.3523,
|
| 751 |
+
"eval_samples_per_second": 1.544,
|
| 752 |
+
"eval_steps_per_second": 0.194,
|
| 753 |
+
"eval_wer": 3.094378545642084,
|
| 754 |
+
"step": 2500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 8.17,
|
| 758 |
+
"grad_norm": 0.0012944688787683845,
|
| 759 |
+
"learning_rate": 4.217142857142858e-06,
|
| 760 |
+
"loss": 0.0,
|
| 761 |
+
"step": 2525
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 8.25,
|
| 765 |
+
"grad_norm": 0.002697590971365571,
|
| 766 |
+
"learning_rate": 4.145714285714286e-06,
|
| 767 |
+
"loss": 0.0,
|
| 768 |
+
"step": 2550
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 8.33,
|
| 772 |
+
"grad_norm": 0.0015665855025872588,
|
| 773 |
+
"learning_rate": 4.074285714285714e-06,
|
| 774 |
+
"loss": 0.0,
|
| 775 |
+
"step": 2575
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 8.41,
|
| 779 |
+
"grad_norm": 0.0023723498452454805,
|
| 780 |
+
"learning_rate": 4.002857142857143e-06,
|
| 781 |
+
"loss": 0.0,
|
| 782 |
+
"step": 2600
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 8.5,
|
| 786 |
+
"grad_norm": 0.0024561272002756596,
|
| 787 |
+
"learning_rate": 3.931428571428571e-06,
|
| 788 |
+
"loss": 0.0,
|
| 789 |
+
"step": 2625
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 8.58,
|
| 793 |
+
"grad_norm": 0.0012809446780011058,
|
| 794 |
+
"learning_rate": 3.86e-06,
|
| 795 |
+
"loss": 0.0,
|
| 796 |
+
"step": 2650
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 8.66,
|
| 800 |
+
"grad_norm": 0.002741657430306077,
|
| 801 |
+
"learning_rate": 3.7885714285714285e-06,
|
| 802 |
+
"loss": 0.0,
|
| 803 |
+
"step": 2675
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 8.74,
|
| 807 |
+
"grad_norm": 0.0022798278369009495,
|
| 808 |
+
"learning_rate": 3.7171428571428575e-06,
|
| 809 |
+
"loss": 0.0,
|
| 810 |
+
"step": 2700
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 8.82,
|
| 814 |
+
"grad_norm": 0.0008324767695739865,
|
| 815 |
+
"learning_rate": 3.6457142857142857e-06,
|
| 816 |
+
"loss": 0.0,
|
| 817 |
+
"step": 2725
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 8.9,
|
| 821 |
+
"grad_norm": 0.0021829677280038595,
|
| 822 |
+
"learning_rate": 3.5742857142857147e-06,
|
| 823 |
+
"loss": 0.0,
|
| 824 |
+
"step": 2750
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 8.98,
|
| 828 |
+
"grad_norm": 0.0010186375584453344,
|
| 829 |
+
"learning_rate": 3.5028571428571433e-06,
|
| 830 |
+
"loss": 0.0,
|
| 831 |
+
"step": 2775
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 9.06,
|
| 835 |
+
"grad_norm": 0.0021447737235575914,
|
| 836 |
+
"learning_rate": 3.431428571428572e-06,
|
| 837 |
+
"loss": 0.0,
|
| 838 |
+
"step": 2800
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 9.14,
|
| 842 |
+
"grad_norm": 0.0014905674615874887,
|
| 843 |
+
"learning_rate": 3.3600000000000004e-06,
|
| 844 |
+
"loss": 0.0,
|
| 845 |
+
"step": 2825
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 9.22,
|
| 849 |
+
"grad_norm": 0.0023917625658214092,
|
| 850 |
+
"learning_rate": 3.2885714285714286e-06,
|
| 851 |
+
"loss": 0.0,
|
| 852 |
+
"step": 2850
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 9.3,
|
| 856 |
+
"grad_norm": 0.0014455171767622232,
|
| 857 |
+
"learning_rate": 3.2171428571428576e-06,
|
| 858 |
+
"loss": 0.0,
|
| 859 |
+
"step": 2875
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 9.39,
|
| 863 |
+
"grad_norm": 0.009605071507394314,
|
| 864 |
+
"learning_rate": 3.1457142857142858e-06,
|
| 865 |
+
"loss": 0.0,
|
| 866 |
+
"step": 2900
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 9.47,
|
| 870 |
+
"grad_norm": 0.0006564524956047535,
|
| 871 |
+
"learning_rate": 3.074285714285715e-06,
|
| 872 |
+
"loss": 0.0,
|
| 873 |
+
"step": 2925
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 9.55,
|
| 877 |
+
"grad_norm": 0.0008893849444575608,
|
| 878 |
+
"learning_rate": 3.002857142857143e-06,
|
| 879 |
+
"loss": 0.0,
|
| 880 |
+
"step": 2950
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 9.63,
|
| 884 |
+
"grad_norm": 0.001537854433991015,
|
| 885 |
+
"learning_rate": 2.9314285714285716e-06,
|
| 886 |
+
"loss": 0.0,
|
| 887 |
+
"step": 2975
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 9.71,
|
| 891 |
+
"grad_norm": 0.0020573874935507774,
|
| 892 |
+
"learning_rate": 2.86e-06,
|
| 893 |
+
"loss": 0.0,
|
| 894 |
+
"step": 3000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 9.71,
|
| 898 |
+
"eval_loss": 0.017799125984311104,
|
| 899 |
+
"eval_runtime": 794.7434,
|
| 900 |
+
"eval_samples_per_second": 1.551,
|
| 901 |
+
"eval_steps_per_second": 0.195,
|
| 902 |
+
"eval_wer": 3.094378545642084,
|
| 903 |
+
"step": 3000
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 9.79,
|
| 907 |
+
"grad_norm": 0.001788494409993291,
|
| 908 |
+
"learning_rate": 2.7885714285714287e-06,
|
| 909 |
+
"loss": 0.0,
|
| 910 |
+
"step": 3025
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 9.87,
|
| 914 |
+
"grad_norm": 0.0009725405252538621,
|
| 915 |
+
"learning_rate": 2.7171428571428577e-06,
|
| 916 |
+
"loss": 0.0,
|
| 917 |
+
"step": 3050
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 9.95,
|
| 921 |
+
"grad_norm": 0.001284493599087,
|
| 922 |
+
"learning_rate": 2.645714285714286e-06,
|
| 923 |
+
"loss": 0.0,
|
| 924 |
+
"step": 3075
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 10.03,
|
| 928 |
+
"grad_norm": 0.0017106755403801799,
|
| 929 |
+
"learning_rate": 2.574285714285715e-06,
|
| 930 |
+
"loss": 0.0,
|
| 931 |
+
"step": 3100
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 10.11,
|
| 935 |
+
"grad_norm": 0.0007921723299659789,
|
| 936 |
+
"learning_rate": 2.502857142857143e-06,
|
| 937 |
+
"loss": 0.0,
|
| 938 |
+
"step": 3125
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 10.19,
|
| 942 |
+
"grad_norm": 0.0018981642788276076,
|
| 943 |
+
"learning_rate": 2.4314285714285717e-06,
|
| 944 |
+
"loss": 0.0,
|
| 945 |
+
"step": 3150
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 10.28,
|
| 949 |
+
"grad_norm": 0.0014530618209391832,
|
| 950 |
+
"learning_rate": 2.3600000000000003e-06,
|
| 951 |
+
"loss": 0.0,
|
| 952 |
+
"step": 3175
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 10.36,
|
| 956 |
+
"grad_norm": 0.0011344915255904198,
|
| 957 |
+
"learning_rate": 2.288571428571429e-06,
|
| 958 |
+
"loss": 0.0,
|
| 959 |
+
"step": 3200
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 10.44,
|
| 963 |
+
"grad_norm": 0.0006109612877480686,
|
| 964 |
+
"learning_rate": 2.2171428571428575e-06,
|
| 965 |
+
"loss": 0.0,
|
| 966 |
+
"step": 3225
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 10.52,
|
| 970 |
+
"grad_norm": 0.0011850837618112564,
|
| 971 |
+
"learning_rate": 2.145714285714286e-06,
|
| 972 |
+
"loss": 0.0,
|
| 973 |
+
"step": 3250
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 10.6,
|
| 977 |
+
"grad_norm": 0.0005283011123538017,
|
| 978 |
+
"learning_rate": 2.0742857142857146e-06,
|
| 979 |
+
"loss": 0.0,
|
| 980 |
+
"step": 3275
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 10.68,
|
| 984 |
+
"grad_norm": 0.0010328389471396804,
|
| 985 |
+
"learning_rate": 2.0028571428571432e-06,
|
| 986 |
+
"loss": 0.0,
|
| 987 |
+
"step": 3300
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 10.76,
|
| 991 |
+
"grad_norm": 0.0009599952027201653,
|
| 992 |
+
"learning_rate": 1.9314285714285714e-06,
|
| 993 |
+
"loss": 0.0,
|
| 994 |
+
"step": 3325
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 10.84,
|
| 998 |
+
"grad_norm": 0.0011755761224776506,
|
| 999 |
+
"learning_rate": 1.8600000000000002e-06,
|
| 1000 |
+
"loss": 0.0,
|
| 1001 |
+
"step": 3350
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 10.92,
|
| 1005 |
+
"grad_norm": 0.0018132536206394434,
|
| 1006 |
+
"learning_rate": 1.7885714285714288e-06,
|
| 1007 |
+
"loss": 0.0,
|
| 1008 |
+
"step": 3375
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 11.0,
|
| 1012 |
+
"grad_norm": 0.0018807642627507448,
|
| 1013 |
+
"learning_rate": 1.7171428571428572e-06,
|
| 1014 |
+
"loss": 0.0,
|
| 1015 |
+
"step": 3400
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 11.08,
|
| 1019 |
+
"grad_norm": 0.003156110644340515,
|
| 1020 |
+
"learning_rate": 1.6457142857142857e-06,
|
| 1021 |
+
"loss": 0.0,
|
| 1022 |
+
"step": 3425
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 11.17,
|
| 1026 |
+
"grad_norm": 0.0010045066010206938,
|
| 1027 |
+
"learning_rate": 1.5742857142857143e-06,
|
| 1028 |
+
"loss": 0.0,
|
| 1029 |
+
"step": 3450
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 11.25,
|
| 1033 |
+
"grad_norm": 0.0008904847782105207,
|
| 1034 |
+
"learning_rate": 1.502857142857143e-06,
|
| 1035 |
+
"loss": 0.0,
|
| 1036 |
+
"step": 3475
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 11.33,
|
| 1040 |
+
"grad_norm": 0.0013107324484735727,
|
| 1041 |
+
"learning_rate": 1.4314285714285717e-06,
|
| 1042 |
+
"loss": 0.0,
|
| 1043 |
+
"step": 3500
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 11.33,
|
| 1047 |
+
"eval_loss": 0.017873059958219528,
|
| 1048 |
+
"eval_runtime": 799.0063,
|
| 1049 |
+
"eval_samples_per_second": 1.543,
|
| 1050 |
+
"eval_steps_per_second": 0.194,
|
| 1051 |
+
"eval_wer": 3.1287605294825513,
|
| 1052 |
+
"step": 3500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 11.41,
|
| 1056 |
+
"grad_norm": 0.0013505747774615884,
|
| 1057 |
+
"learning_rate": 1.3600000000000001e-06,
|
| 1058 |
+
"loss": 0.0,
|
| 1059 |
+
"step": 3525
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 11.49,
|
| 1063 |
+
"grad_norm": 0.0013679265975952148,
|
| 1064 |
+
"learning_rate": 1.2885714285714287e-06,
|
| 1065 |
+
"loss": 0.0,
|
| 1066 |
+
"step": 3550
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 11.57,
|
| 1070 |
+
"grad_norm": 0.0021877065300941467,
|
| 1071 |
+
"learning_rate": 1.2171428571428573e-06,
|
| 1072 |
+
"loss": 0.0,
|
| 1073 |
+
"step": 3575
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 11.65,
|
| 1077 |
+
"grad_norm": 0.0011414324399083853,
|
| 1078 |
+
"learning_rate": 1.1457142857142859e-06,
|
| 1079 |
+
"loss": 0.0,
|
| 1080 |
+
"step": 3600
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 11.73,
|
| 1084 |
+
"grad_norm": 0.0008932927739806473,
|
| 1085 |
+
"learning_rate": 1.0742857142857145e-06,
|
| 1086 |
+
"loss": 0.0,
|
| 1087 |
+
"step": 3625
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 11.81,
|
| 1091 |
+
"grad_norm": 0.0016671326011419296,
|
| 1092 |
+
"learning_rate": 1.0028571428571428e-06,
|
| 1093 |
+
"loss": 0.0,
|
| 1094 |
+
"step": 3650
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 11.89,
|
| 1098 |
+
"grad_norm": 0.0011023151455447078,
|
| 1099 |
+
"learning_rate": 9.314285714285714e-07,
|
| 1100 |
+
"loss": 0.0,
|
| 1101 |
+
"step": 3675
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 11.97,
|
| 1105 |
+
"grad_norm": 0.0008648216025903821,
|
| 1106 |
+
"learning_rate": 8.6e-07,
|
| 1107 |
+
"loss": 0.0,
|
| 1108 |
+
"step": 3700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 12.06,
|
| 1112 |
+
"grad_norm": 0.003709783311933279,
|
| 1113 |
+
"learning_rate": 7.885714285714287e-07,
|
| 1114 |
+
"loss": 0.0,
|
| 1115 |
+
"step": 3725
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 12.14,
|
| 1119 |
+
"grad_norm": 0.0005944801960140467,
|
| 1120 |
+
"learning_rate": 7.171428571428572e-07,
|
| 1121 |
+
"loss": 0.0,
|
| 1122 |
+
"step": 3750
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 12.22,
|
| 1126 |
+
"grad_norm": 0.0010830622632056475,
|
| 1127 |
+
"learning_rate": 6.457142857142858e-07,
|
| 1128 |
+
"loss": 0.0,
|
| 1129 |
+
"step": 3775
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 12.3,
|
| 1133 |
+
"grad_norm": 0.0018471087096258998,
|
| 1134 |
+
"learning_rate": 5.742857142857143e-07,
|
| 1135 |
+
"loss": 0.0,
|
| 1136 |
+
"step": 3800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 12.38,
|
| 1140 |
+
"grad_norm": 0.0018315908964723349,
|
| 1141 |
+
"learning_rate": 5.028571428571429e-07,
|
| 1142 |
+
"loss": 0.0,
|
| 1143 |
+
"step": 3825
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 12.46,
|
| 1147 |
+
"grad_norm": 0.0010556795168668032,
|
| 1148 |
+
"learning_rate": 4.3142857142857146e-07,
|
| 1149 |
+
"loss": 0.0,
|
| 1150 |
+
"step": 3850
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 12.54,
|
| 1154 |
+
"grad_norm": 0.001286807470023632,
|
| 1155 |
+
"learning_rate": 3.6e-07,
|
| 1156 |
+
"loss": 0.0,
|
| 1157 |
+
"step": 3875
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 12.62,
|
| 1161 |
+
"grad_norm": 0.0025280534755438566,
|
| 1162 |
+
"learning_rate": 2.885714285714286e-07,
|
| 1163 |
+
"loss": 0.0,
|
| 1164 |
+
"step": 3900
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 12.7,
|
| 1168 |
+
"grad_norm": 0.0008814557804726064,
|
| 1169 |
+
"learning_rate": 2.1714285714285715e-07,
|
| 1170 |
+
"loss": 0.0,
|
| 1171 |
+
"step": 3925
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 12.78,
|
| 1175 |
+
"grad_norm": 0.0007184511632658541,
|
| 1176 |
+
"learning_rate": 1.4571428571428574e-07,
|
| 1177 |
+
"loss": 0.0,
|
| 1178 |
+
"step": 3950
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 12.86,
|
| 1182 |
+
"grad_norm": 0.0024200354237109423,
|
| 1183 |
+
"learning_rate": 7.428571428571429e-08,
|
| 1184 |
+
"loss": 0.0,
|
| 1185 |
+
"step": 3975
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 12.94,
|
| 1189 |
+
"grad_norm": 0.0013352702371776104,
|
| 1190 |
+
"learning_rate": 2.8571428571428576e-09,
|
| 1191 |
+
"loss": 0.0,
|
| 1192 |
+
"step": 4000
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 12.94,
|
| 1196 |
+
"eval_loss": 0.017900340259075165,
|
| 1197 |
+
"eval_runtime": 798.968,
|
| 1198 |
+
"eval_samples_per_second": 1.543,
|
| 1199 |
+
"eval_steps_per_second": 0.194,
|
| 1200 |
+
"eval_wer": 3.1287605294825513,
|
| 1201 |
+
"step": 4000
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 13.03,
|
| 1205 |
+
"grad_norm": 0.0006061609601601958,
|
| 1206 |
+
"learning_rate": 5.301333333333334e-06,
|
| 1207 |
+
"loss": 0.0,
|
| 1208 |
+
"step": 4025
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 13.11,
|
| 1212 |
+
"grad_norm": 0.000903214851859957,
|
| 1213 |
+
"learning_rate": 5.268000000000001e-06,
|
| 1214 |
+
"loss": 0.0,
|
| 1215 |
+
"step": 4050
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 13.19,
|
| 1219 |
+
"grad_norm": 0.0009095696732401848,
|
| 1220 |
+
"learning_rate": 5.234666666666667e-06,
|
| 1221 |
+
"loss": 0.0,
|
| 1222 |
+
"step": 4075
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 13.27,
|
| 1226 |
+
"grad_norm": 0.0009569237590767443,
|
| 1227 |
+
"learning_rate": 5.201333333333334e-06,
|
| 1228 |
+
"loss": 0.0,
|
| 1229 |
+
"step": 4100
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 13.35,
|
| 1233 |
+
"grad_norm": 0.0013346378691494465,
|
| 1234 |
+
"learning_rate": 5.168000000000001e-06,
|
| 1235 |
+
"loss": 0.0,
|
| 1236 |
+
"step": 4125
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 13.43,
|
| 1240 |
+
"grad_norm": 0.0007105122786015272,
|
| 1241 |
+
"learning_rate": 5.134666666666667e-06,
|
| 1242 |
+
"loss": 0.0,
|
| 1243 |
+
"step": 4150
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 13.51,
|
| 1247 |
+
"grad_norm": 0.001324965967796743,
|
| 1248 |
+
"learning_rate": 5.101333333333334e-06,
|
| 1249 |
+
"loss": 0.0,
|
| 1250 |
+
"step": 4175
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 13.59,
|
| 1254 |
+
"grad_norm": 0.002936706179752946,
|
| 1255 |
+
"learning_rate": 5.0680000000000004e-06,
|
| 1256 |
+
"loss": 0.0,
|
| 1257 |
+
"step": 4200
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 13.67,
|
| 1261 |
+
"grad_norm": 0.0006480725132860243,
|
| 1262 |
+
"learning_rate": 5.034666666666667e-06,
|
| 1263 |
+
"loss": 0.0,
|
| 1264 |
+
"step": 4225
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 13.75,
|
| 1268 |
+
"grad_norm": 0.0008248516824096441,
|
| 1269 |
+
"learning_rate": 5.001333333333333e-06,
|
| 1270 |
+
"loss": 0.0,
|
| 1271 |
+
"step": 4250
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 13.83,
|
| 1275 |
+
"grad_norm": 0.000976664014160633,
|
| 1276 |
+
"learning_rate": 4.9680000000000005e-06,
|
| 1277 |
+
"loss": 0.0,
|
| 1278 |
+
"step": 4275
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 13.92,
|
| 1282 |
+
"grad_norm": 0.0007901455392129719,
|
| 1283 |
+
"learning_rate": 4.934666666666667e-06,
|
| 1284 |
+
"loss": 0.0,
|
| 1285 |
+
"step": 4300
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 14.0,
|
| 1289 |
+
"grad_norm": 0.0008691786788403988,
|
| 1290 |
+
"learning_rate": 4.901333333333333e-06,
|
| 1291 |
+
"loss": 0.0,
|
| 1292 |
+
"step": 4325
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 14.08,
|
| 1296 |
+
"grad_norm": 0.00039108938653953373,
|
| 1297 |
+
"learning_rate": 4.868000000000001e-06,
|
| 1298 |
+
"loss": 0.0,
|
| 1299 |
+
"step": 4350
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 14.16,
|
| 1303 |
+
"grad_norm": 0.000672497262712568,
|
| 1304 |
+
"learning_rate": 4.834666666666667e-06,
|
| 1305 |
+
"loss": 0.0,
|
| 1306 |
+
"step": 4375
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 14.24,
|
| 1310 |
+
"grad_norm": 0.0006290368619374931,
|
| 1311 |
+
"learning_rate": 4.801333333333334e-06,
|
| 1312 |
+
"loss": 0.0,
|
| 1313 |
+
"step": 4400
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 14.32,
|
| 1317 |
+
"grad_norm": 0.0012541558826342225,
|
| 1318 |
+
"learning_rate": 4.768000000000001e-06,
|
| 1319 |
+
"loss": 0.0,
|
| 1320 |
+
"step": 4425
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 14.4,
|
| 1324 |
+
"grad_norm": 0.0015149698592722416,
|
| 1325 |
+
"learning_rate": 4.734666666666667e-06,
|
| 1326 |
+
"loss": 0.0,
|
| 1327 |
+
"step": 4450
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 14.48,
|
| 1331 |
+
"grad_norm": 0.001397727057337761,
|
| 1332 |
+
"learning_rate": 4.701333333333334e-06,
|
| 1333 |
+
"loss": 0.0,
|
| 1334 |
+
"step": 4475
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 14.56,
|
| 1338 |
+
"grad_norm": 0.001000353367999196,
|
| 1339 |
+
"learning_rate": 4.668e-06,
|
| 1340 |
+
"loss": 0.0,
|
| 1341 |
+
"step": 4500
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 14.56,
|
| 1345 |
+
"eval_loss": 0.018149884417653084,
|
| 1346 |
+
"eval_runtime": 806.023,
|
| 1347 |
+
"eval_samples_per_second": 1.53,
|
| 1348 |
+
"eval_steps_per_second": 0.192,
|
| 1349 |
+
"eval_wer": 2.888086642599278,
|
| 1350 |
+
"step": 4500
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 14.64,
|
| 1354 |
+
"grad_norm": 0.0004112539754714817,
|
| 1355 |
+
"learning_rate": 4.634666666666667e-06,
|
| 1356 |
+
"loss": 0.0,
|
| 1357 |
+
"step": 4525
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 14.72,
|
| 1361 |
+
"grad_norm": 0.00042669521644711494,
|
| 1362 |
+
"learning_rate": 4.601333333333334e-06,
|
| 1363 |
+
"loss": 0.0,
|
| 1364 |
+
"step": 4550
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 14.81,
|
| 1368 |
+
"grad_norm": 0.0005972330691292882,
|
| 1369 |
+
"learning_rate": 4.568e-06,
|
| 1370 |
+
"loss": 0.0,
|
| 1371 |
+
"step": 4575
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 14.89,
|
| 1375 |
+
"grad_norm": 0.000703622295986861,
|
| 1376 |
+
"learning_rate": 4.534666666666667e-06,
|
| 1377 |
+
"loss": 0.0,
|
| 1378 |
+
"step": 4600
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 14.97,
|
| 1382 |
+
"grad_norm": 0.0008054905338212848,
|
| 1383 |
+
"learning_rate": 4.501333333333334e-06,
|
| 1384 |
+
"loss": 0.0,
|
| 1385 |
+
"step": 4625
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 15.05,
|
| 1389 |
+
"grad_norm": 0.0006107372464612126,
|
| 1390 |
+
"learning_rate": 4.468e-06,
|
| 1391 |
+
"loss": 0.0,
|
| 1392 |
+
"step": 4650
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 15.13,
|
| 1396 |
+
"grad_norm": 0.0011443241965025663,
|
| 1397 |
+
"learning_rate": 4.4346666666666675e-06,
|
| 1398 |
+
"loss": 0.0,
|
| 1399 |
+
"step": 4675
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 15.21,
|
| 1403 |
+
"grad_norm": 0.0007337784627452493,
|
| 1404 |
+
"learning_rate": 4.401333333333334e-06,
|
| 1405 |
+
"loss": 0.0,
|
| 1406 |
+
"step": 4700
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 15.29,
|
| 1410 |
+
"grad_norm": 0.0005867818254046142,
|
| 1411 |
+
"learning_rate": 4.368e-06,
|
| 1412 |
+
"loss": 0.0,
|
| 1413 |
+
"step": 4725
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 15.37,
|
| 1417 |
+
"grad_norm": 0.00047605938743799925,
|
| 1418 |
+
"learning_rate": 4.334666666666667e-06,
|
| 1419 |
+
"loss": 0.0,
|
| 1420 |
+
"step": 4750
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 15.45,
|
| 1424 |
+
"grad_norm": 0.0007535215700045228,
|
| 1425 |
+
"learning_rate": 4.301333333333333e-06,
|
| 1426 |
+
"loss": 0.0,
|
| 1427 |
+
"step": 4775
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 15.53,
|
| 1431 |
+
"grad_norm": 0.0004895786987617612,
|
| 1432 |
+
"learning_rate": 4.2680000000000005e-06,
|
| 1433 |
+
"loss": 0.0,
|
| 1434 |
+
"step": 4800
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 15.61,
|
| 1438 |
+
"grad_norm": 0.0005502207204699516,
|
| 1439 |
+
"learning_rate": 4.234666666666667e-06,
|
| 1440 |
+
"loss": 0.0,
|
| 1441 |
+
"step": 4825
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 15.7,
|
| 1445 |
+
"grad_norm": 0.0008273226558230817,
|
| 1446 |
+
"learning_rate": 4.201333333333334e-06,
|
| 1447 |
+
"loss": 0.0,
|
| 1448 |
+
"step": 4850
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 15.78,
|
| 1452 |
+
"grad_norm": 0.0008853155304677784,
|
| 1453 |
+
"learning_rate": 4.168000000000001e-06,
|
| 1454 |
+
"loss": 0.0,
|
| 1455 |
+
"step": 4875
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 15.86,
|
| 1459 |
+
"grad_norm": 0.0008551743812859058,
|
| 1460 |
+
"learning_rate": 4.134666666666667e-06,
|
| 1461 |
+
"loss": 0.0,
|
| 1462 |
+
"step": 4900
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 15.94,
|
| 1466 |
+
"grad_norm": 0.0012048580683767796,
|
| 1467 |
+
"learning_rate": 4.1013333333333334e-06,
|
| 1468 |
+
"loss": 0.0,
|
| 1469 |
+
"step": 4925
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 16.02,
|
| 1473 |
+
"grad_norm": 0.00043612357694655657,
|
| 1474 |
+
"learning_rate": 4.068000000000001e-06,
|
| 1475 |
+
"loss": 0.0,
|
| 1476 |
+
"step": 4950
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 16.1,
|
| 1480 |
+
"grad_norm": 0.0005169506184756756,
|
| 1481 |
+
"learning_rate": 4.034666666666667e-06,
|
| 1482 |
+
"loss": 0.0,
|
| 1483 |
+
"step": 4975
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 16.18,
|
| 1487 |
+
"grad_norm": 0.000585487752687186,
|
| 1488 |
+
"learning_rate": 4.0013333333333335e-06,
|
| 1489 |
+
"loss": 0.0,
|
| 1490 |
+
"step": 5000
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 16.18,
|
| 1494 |
+
"eval_loss": 0.018417511135339737,
|
| 1495 |
+
"eval_runtime": 808.2166,
|
| 1496 |
+
"eval_samples_per_second": 1.526,
|
| 1497 |
+
"eval_steps_per_second": 0.192,
|
| 1498 |
+
"eval_wer": 2.9224686264397457,
|
| 1499 |
+
"step": 5000
|
| 1500 |
+
}
|
| 1501 |
+
],
|
| 1502 |
+
"logging_steps": 25,
|
| 1503 |
+
"max_steps": 8000,
|
| 1504 |
+
"num_input_tokens_seen": 0,
|
| 1505 |
+
"num_train_epochs": 26,
|
| 1506 |
+
"save_steps": 1000,
|
| 1507 |
+
"total_flos": 2.303142362873856e+19,
|
| 1508 |
+
"train_batch_size": 16,
|
| 1509 |
+
"trial_name": null,
|
| 1510 |
+
"trial_params": null
|
| 1511 |
+
}
|
checkpoint-5000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81fdd69ed4230c843750246ce946cd9653ab2226cbd46ba20a262fd6ffd7c78b
|
| 3 |
+
size 5112
|