Upload 9 files
Browse files- config.json +101 -5
- generation_config.json +1 -28
- model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +625 -1209
- training_args.bin +2 -2
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"apply_spec_augment": false,
|
|
@@ -25,7 +25,12 @@
|
|
| 25 |
"encoder_layerdrop": 0.0,
|
| 26 |
"encoder_layers": 6,
|
| 27 |
"eos_token_id": 50256,
|
| 28 |
-
"forced_decoder_ids":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"init_std": 0.02,
|
| 30 |
"is_encoder_decoder": true,
|
| 31 |
"mask_feature_length": 10,
|
|
@@ -43,10 +48,101 @@
|
|
| 43 |
"num_mel_bins": 80,
|
| 44 |
"pad_token_id": 50256,
|
| 45 |
"scale_embedding": false,
|
| 46 |
-
"suppress_tokens": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
"torch_dtype": "float32",
|
| 48 |
-
"transformers_version": "4.
|
| 49 |
-
"use_cache":
|
| 50 |
"use_weighted_layer_sum": false,
|
| 51 |
"vocab_size": 51864
|
| 52 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/kaggle/working/whisper-ft-2/checkpoint-600/model.safetensors",
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"apply_spec_augment": false,
|
|
|
|
| 25 |
"encoder_layerdrop": 0.0,
|
| 26 |
"encoder_layers": 6,
|
| 27 |
"eos_token_id": 50256,
|
| 28 |
+
"forced_decoder_ids": [
|
| 29 |
+
[
|
| 30 |
+
1,
|
| 31 |
+
50362
|
| 32 |
+
]
|
| 33 |
+
],
|
| 34 |
"init_std": 0.02,
|
| 35 |
"is_encoder_decoder": true,
|
| 36 |
"mask_feature_length": 10,
|
|
|
|
| 48 |
"num_mel_bins": 80,
|
| 49 |
"pad_token_id": 50256,
|
| 50 |
"scale_embedding": false,
|
| 51 |
+
"suppress_tokens": [
|
| 52 |
+
1,
|
| 53 |
+
2,
|
| 54 |
+
7,
|
| 55 |
+
8,
|
| 56 |
+
9,
|
| 57 |
+
10,
|
| 58 |
+
14,
|
| 59 |
+
25,
|
| 60 |
+
26,
|
| 61 |
+
27,
|
| 62 |
+
28,
|
| 63 |
+
29,
|
| 64 |
+
31,
|
| 65 |
+
58,
|
| 66 |
+
59,
|
| 67 |
+
60,
|
| 68 |
+
61,
|
| 69 |
+
62,
|
| 70 |
+
63,
|
| 71 |
+
90,
|
| 72 |
+
91,
|
| 73 |
+
92,
|
| 74 |
+
93,
|
| 75 |
+
357,
|
| 76 |
+
366,
|
| 77 |
+
438,
|
| 78 |
+
532,
|
| 79 |
+
685,
|
| 80 |
+
705,
|
| 81 |
+
796,
|
| 82 |
+
930,
|
| 83 |
+
1058,
|
| 84 |
+
1220,
|
| 85 |
+
1267,
|
| 86 |
+
1279,
|
| 87 |
+
1303,
|
| 88 |
+
1343,
|
| 89 |
+
1377,
|
| 90 |
+
1391,
|
| 91 |
+
1635,
|
| 92 |
+
1782,
|
| 93 |
+
1875,
|
| 94 |
+
2162,
|
| 95 |
+
2361,
|
| 96 |
+
2488,
|
| 97 |
+
3467,
|
| 98 |
+
4008,
|
| 99 |
+
4211,
|
| 100 |
+
4600,
|
| 101 |
+
4808,
|
| 102 |
+
5299,
|
| 103 |
+
5855,
|
| 104 |
+
6329,
|
| 105 |
+
7203,
|
| 106 |
+
9609,
|
| 107 |
+
9959,
|
| 108 |
+
10563,
|
| 109 |
+
10786,
|
| 110 |
+
11420,
|
| 111 |
+
11709,
|
| 112 |
+
11907,
|
| 113 |
+
13163,
|
| 114 |
+
13697,
|
| 115 |
+
13700,
|
| 116 |
+
14808,
|
| 117 |
+
15306,
|
| 118 |
+
16410,
|
| 119 |
+
16791,
|
| 120 |
+
17992,
|
| 121 |
+
19203,
|
| 122 |
+
19510,
|
| 123 |
+
20724,
|
| 124 |
+
22305,
|
| 125 |
+
22935,
|
| 126 |
+
27007,
|
| 127 |
+
30109,
|
| 128 |
+
30420,
|
| 129 |
+
33409,
|
| 130 |
+
34949,
|
| 131 |
+
40283,
|
| 132 |
+
40493,
|
| 133 |
+
40549,
|
| 134 |
+
47282,
|
| 135 |
+
49146,
|
| 136 |
+
50257,
|
| 137 |
+
50357,
|
| 138 |
+
50358,
|
| 139 |
+
50359,
|
| 140 |
+
50360,
|
| 141 |
+
50361
|
| 142 |
+
],
|
| 143 |
"torch_dtype": "float32",
|
| 144 |
+
"transformers_version": "4.42.3",
|
| 145 |
+
"use_cache": true,
|
| 146 |
"use_weighted_layer_sum": false,
|
| 147 |
"vocab_size": 51864
|
| 148 |
}
|
generation_config.json
CHANGED
|
@@ -1,26 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"alignment_heads": [
|
| 3 |
-
[
|
| 4 |
-
3,
|
| 5 |
-
3
|
| 6 |
-
],
|
| 7 |
-
[
|
| 8 |
-
4,
|
| 9 |
-
7
|
| 10 |
-
],
|
| 11 |
-
[
|
| 12 |
-
5,
|
| 13 |
-
1
|
| 14 |
-
],
|
| 15 |
-
[
|
| 16 |
-
5,
|
| 17 |
-
5
|
| 18 |
-
],
|
| 19 |
-
[
|
| 20 |
-
5,
|
| 21 |
-
7
|
| 22 |
-
]
|
| 23 |
-
],
|
| 24 |
"begin_suppress_tokens": [
|
| 25 |
220,
|
| 26 |
50256
|
|
@@ -34,13 +12,8 @@
|
|
| 34 |
50362
|
| 35 |
]
|
| 36 |
],
|
| 37 |
-
"is_multilingual": false,
|
| 38 |
-
"max_initial_timestamp_index": 50,
|
| 39 |
"max_length": 448,
|
| 40 |
-
"no_timestamps_token_id": 50362,
|
| 41 |
"pad_token_id": 50256,
|
| 42 |
-
"prev_sot_token_id": 50360,
|
| 43 |
-
"return_timestamps": false,
|
| 44 |
"suppress_tokens": [
|
| 45 |
1,
|
| 46 |
2,
|
|
@@ -133,5 +106,5 @@
|
|
| 133 |
50360,
|
| 134 |
50361
|
| 135 |
],
|
| 136 |
-
"transformers_version": "4.
|
| 137 |
}
|
|
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"begin_suppress_tokens": [
|
| 3 |
220,
|
| 4 |
50256
|
|
|
|
| 12 |
50362
|
| 13 |
]
|
| 14 |
],
|
|
|
|
|
|
|
| 15 |
"max_length": 448,
|
|
|
|
| 16 |
"pad_token_id": 50256,
|
|
|
|
|
|
|
| 17 |
"suppress_tokens": [
|
| 18 |
1,
|
| 19 |
2,
|
|
|
|
| 106 |
50360,
|
| 107 |
50361
|
| 108 |
],
|
| 109 |
+
"transformers_version": "4.42.3"
|
| 110 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 290401888
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:22aa2c1a7ff282ac49e8b7b35970dc582ab111868362fb5d325524b1e181646a
|
| 3 |
size 290401888
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 574807418
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e858eb0107d3a7caf2bce9b6dcbae6286b6e0670211dd4e01d9772be12056a79
|
| 3 |
size 574807418
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49f603c8d0789e2771e72e17a35d09664975138faec34bc1a21798b82a6ebaa3
|
| 3 |
size 14244
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78f763231c165c2571b07da4d26c7965824aff598ab2cc82420929859dcb97f0
|
| 3 |
size 1064
|
trainer_state.json
CHANGED
|
@@ -1,1466 +1,882 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./whisper-ft-2/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 0.
|
| 16 |
-
"step":
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.01,
|
| 20 |
-
"grad_norm": 5.579243183135986,
|
| 21 |
-
"learning_rate": 9.96376811594203e-06,
|
| 22 |
-
"loss": 0.1739,
|
| 23 |
-
"step": 50
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.01,
|
| 27 |
-
"grad_norm": 4.500240325927734,
|
| 28 |
-
"learning_rate": 9.913446054750404e-06,
|
| 29 |
-
"loss": 0.1795,
|
| 30 |
-
"step": 75
|
| 31 |
-
},
|
| 32 |
-
{
|
| 33 |
-
"epoch": 0.02,
|
| 34 |
-
"grad_norm": 2.341679811477661,
|
| 35 |
-
"learning_rate": 9.863123993558778e-06,
|
| 36 |
-
"loss": 0.1617,
|
| 37 |
-
"step": 100
|
| 38 |
-
},
|
| 39 |
-
{
|
| 40 |
-
"epoch": 0.03,
|
| 41 |
-
"grad_norm": 5.624894618988037,
|
| 42 |
-
"learning_rate": 9.81280193236715e-06,
|
| 43 |
-
"loss": 0.1669,
|
| 44 |
-
"step": 125
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.03,
|
| 48 |
-
"grad_norm": 6.136338710784912,
|
| 49 |
-
"learning_rate": 9.762479871175523e-06,
|
| 50 |
-
"loss": 0.1635,
|
| 51 |
-
"step": 150
|
| 52 |
-
},
|
| 53 |
-
{
|
| 54 |
-
"epoch": 0.04,
|
| 55 |
-
"grad_norm": 4.855004787445068,
|
| 56 |
-
"learning_rate": 9.712157809983898e-06,
|
| 57 |
-
"loss": 0.1611,
|
| 58 |
-
"step": 175
|
| 59 |
-
},
|
| 60 |
-
{
|
| 61 |
-
"epoch": 0.04,
|
| 62 |
-
"grad_norm": 6.826947212219238,
|
| 63 |
-
"learning_rate": 9.66183574879227e-06,
|
| 64 |
-
"loss": 0.1661,
|
| 65 |
-
"step": 200
|
| 66 |
-
},
|
| 67 |
-
{
|
| 68 |
-
"epoch": 0.04,
|
| 69 |
-
"grad_norm": 4.076170921325684,
|
| 70 |
-
"learning_rate": 9.611513687600645e-06,
|
| 71 |
-
"loss": 0.1754,
|
| 72 |
-
"step": 225
|
| 73 |
-
},
|
| 74 |
-
{
|
| 75 |
-
"epoch": 0.05,
|
| 76 |
-
"grad_norm": 7.027714729309082,
|
| 77 |
-
"learning_rate": 9.56119162640902e-06,
|
| 78 |
-
"loss": 0.2,
|
| 79 |
-
"step": 250
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.06,
|
| 83 |
-
"grad_norm": 4.252183437347412,
|
| 84 |
-
"learning_rate": 9.510869565217392e-06,
|
| 85 |
-
"loss": 0.1827,
|
| 86 |
-
"step": 275
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
-
"step":
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.07,
|
| 97 |
-
"grad_norm": 5.446494102478027,
|
| 98 |
-
"learning_rate": 9.41022544283414e-06,
|
| 99 |
-
"loss": 0.1949,
|
| 100 |
-
"step": 325
|
| 101 |
-
},
|
| 102 |
-
{
|
| 103 |
-
"epoch": 0.07,
|
| 104 |
-
"grad_norm": 4.943406581878662,
|
| 105 |
-
"learning_rate": 9.359903381642514e-06,
|
| 106 |
-
"loss": 0.1732,
|
| 107 |
-
"step": 350
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 0.07,
|
| 111 |
-
"grad_norm": 6.656269073486328,
|
| 112 |
-
"learning_rate": 9.309581320450886e-06,
|
| 113 |
-
"loss": 0.1742,
|
| 114 |
-
"step": 375
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 0.08,
|
| 118 |
-
"grad_norm": 4.511006832122803,
|
| 119 |
-
"learning_rate": 9.25925925925926e-06,
|
| 120 |
-
"loss": 0.163,
|
| 121 |
-
"step": 400
|
| 122 |
-
},
|
| 123 |
-
{
|
| 124 |
-
"epoch": 0.09,
|
| 125 |
-
"grad_norm": 4.760077953338623,
|
| 126 |
-
"learning_rate": 9.208937198067634e-06,
|
| 127 |
-
"loss": 0.1564,
|
| 128 |
-
"step": 425
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 0.09,
|
| 132 |
-
"grad_norm": 5.462475299835205,
|
| 133 |
-
"learning_rate": 9.160628019323673e-06,
|
| 134 |
-
"loss": 0.1381,
|
| 135 |
-
"step": 450
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.1,
|
| 139 |
-
"grad_norm":
|
| 140 |
-
"learning_rate": 9.
|
| 141 |
-
"loss": 0.
|
| 142 |
-
"step":
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm":
|
| 147 |
-
"learning_rate": 9.
|
| 148 |
-
"loss": 0.
|
| 149 |
-
"step":
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm":
|
| 154 |
-
"learning_rate": 9.
|
| 155 |
-
"loss": 0.
|
| 156 |
-
"step":
|
| 157 |
-
},
|
| 158 |
-
{
|
| 159 |
-
"epoch": 0.11,
|
| 160 |
-
"grad_norm": 5.756856918334961,
|
| 161 |
-
"learning_rate": 8.959339774557167e-06,
|
| 162 |
-
"loss": 0.1043,
|
| 163 |
-
"step": 550
|
| 164 |
-
},
|
| 165 |
-
{
|
| 166 |
-
"epoch": 0.12,
|
| 167 |
-
"grad_norm": 4.615521430969238,
|
| 168 |
-
"learning_rate": 8.90901771336554e-06,
|
| 169 |
-
"loss": 0.1617,
|
| 170 |
-
"step": 575
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 0.12,
|
| 174 |
-
"grad_norm": 7.172885894775391,
|
| 175 |
-
"learning_rate": 8.858695652173914e-06,
|
| 176 |
-
"loss": 0.166,
|
| 177 |
-
"step": 600
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 0.12,
|
| 181 |
-
"grad_norm": 4.559473514556885,
|
| 182 |
-
"learning_rate": 8.808373590982288e-06,
|
| 183 |
-
"loss": 0.1274,
|
| 184 |
-
"step": 625
|
| 185 |
-
},
|
| 186 |
-
{
|
| 187 |
-
"epoch": 0.13,
|
| 188 |
-
"grad_norm": 4.201319694519043,
|
| 189 |
-
"learning_rate": 8.758051529790661e-06,
|
| 190 |
-
"loss": 0.1382,
|
| 191 |
-
"step": 650
|
| 192 |
-
},
|
| 193 |
-
{
|
| 194 |
-
"epoch": 0.14,
|
| 195 |
-
"grad_norm": 4.814964771270752,
|
| 196 |
-
"learning_rate": 8.707729468599034e-06,
|
| 197 |
-
"loss": 0.1448,
|
| 198 |
-
"step": 675
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"epoch": 0.14,
|
| 202 |
-
"grad_norm": 5.314846992492676,
|
| 203 |
-
"learning_rate": 8.657407407407408e-06,
|
| 204 |
-
"loss": 0.1513,
|
| 205 |
-
"step": 700
|
| 206 |
-
},
|
| 207 |
-
{
|
| 208 |
-
"epoch": 0.14,
|
| 209 |
-
"grad_norm": 3.898376941680908,
|
| 210 |
-
"learning_rate": 8.607085346215783e-06,
|
| 211 |
-
"loss": 0.1634,
|
| 212 |
-
"step": 725
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 0.15,
|
| 216 |
-
"grad_norm": 5.028552055358887,
|
| 217 |
-
"learning_rate": 8.556763285024155e-06,
|
| 218 |
-
"loss": 0.1584,
|
| 219 |
-
"step": 750
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 0.15,
|
| 223 |
-
"grad_norm": 5.497000694274902,
|
| 224 |
-
"learning_rate": 8.506441223832528e-06,
|
| 225 |
-
"loss": 0.1642,
|
| 226 |
-
"step": 775
|
| 227 |
-
},
|
| 228 |
-
{
|
| 229 |
-
"epoch": 0.16,
|
| 230 |
-
"grad_norm": 4.642214298248291,
|
| 231 |
-
"learning_rate": 8.456119162640902e-06,
|
| 232 |
-
"loss": 0.174,
|
| 233 |
-
"step": 800
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 0.17,
|
| 237 |
-
"grad_norm": 4.980218887329102,
|
| 238 |
-
"learning_rate": 8.405797101449275e-06,
|
| 239 |
-
"loss": 0.1885,
|
| 240 |
-
"step": 825
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 0.17,
|
| 244 |
-
"grad_norm": 4.9581618309021,
|
| 245 |
-
"learning_rate": 8.35547504025765e-06,
|
| 246 |
-
"loss": 0.1928,
|
| 247 |
-
"step": 850
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 0.17,
|
| 251 |
-
"grad_norm": 4.621245861053467,
|
| 252 |
-
"learning_rate": 8.305152979066024e-06,
|
| 253 |
-
"loss": 0.1629,
|
| 254 |
-
"step": 875
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 0.18,
|
| 258 |
-
"grad_norm": 4.990960121154785,
|
| 259 |
-
"learning_rate": 8.254830917874397e-06,
|
| 260 |
-
"loss": 0.206,
|
| 261 |
-
"step": 900
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 0.18,
|
| 265 |
-
"grad_norm": 4.905004501342773,
|
| 266 |
-
"learning_rate": 8.20450885668277e-06,
|
| 267 |
-
"loss": 0.232,
|
| 268 |
-
"step": 925
|
| 269 |
-
},
|
| 270 |
-
{
|
| 271 |
-
"epoch": 0.19,
|
| 272 |
-
"grad_norm": 6.192000865936279,
|
| 273 |
-
"learning_rate": 8.154186795491144e-06,
|
| 274 |
-
"loss": 0.2406,
|
| 275 |
-
"step": 950
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 0.2,
|
| 279 |
-
"grad_norm": 4.241088390350342,
|
| 280 |
-
"learning_rate": 8.103864734299518e-06,
|
| 281 |
-
"loss": 0.206,
|
| 282 |
-
"step": 975
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 0.2,
|
| 286 |
-
"grad_norm": 5.220916271209717,
|
| 287 |
-
"learning_rate": 8.053542673107891e-06,
|
| 288 |
-
"loss": 0.2176,
|
| 289 |
-
"step": 1000
|
| 290 |
-
},
|
| 291 |
-
{
|
| 292 |
-
"epoch": 0.2,
|
| 293 |
-
"eval_loss": 0.3194602429866791,
|
| 294 |
-
"eval_runtime": 281.282,
|
| 295 |
-
"eval_samples_per_second": 3.555,
|
| 296 |
-
"eval_steps_per_second": 0.444,
|
| 297 |
-
"eval_wer": 61.69442848937392,
|
| 298 |
-
"step": 1000
|
| 299 |
},
|
| 300 |
{
|
| 301 |
"epoch": 0.2,
|
| 302 |
-
"grad_norm":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 0.
|
| 305 |
-
"step":
|
| 306 |
-
},
|
| 307 |
-
{
|
| 308 |
-
"epoch": 0.21,
|
| 309 |
-
"grad_norm": 6.92851448059082,
|
| 310 |
-
"learning_rate": 7.952898550724638e-06,
|
| 311 |
-
"loss": 0.1915,
|
| 312 |
-
"step": 1050
|
| 313 |
-
},
|
| 314 |
-
{
|
| 315 |
-
"epoch": 0.21,
|
| 316 |
-
"grad_norm": 6.4899725914001465,
|
| 317 |
-
"learning_rate": 7.90257648953301e-06,
|
| 318 |
-
"loss": 0.2366,
|
| 319 |
-
"step": 1075
|
| 320 |
-
},
|
| 321 |
-
{
|
| 322 |
-
"epoch": 0.22,
|
| 323 |
-
"grad_norm": 6.207301139831543,
|
| 324 |
-
"learning_rate": 7.852254428341385e-06,
|
| 325 |
-
"loss": 0.2334,
|
| 326 |
-
"step": 1100
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 0.23,
|
| 330 |
-
"grad_norm": 7.392577648162842,
|
| 331 |
-
"learning_rate": 7.80193236714976e-06,
|
| 332 |
-
"loss": 0.2388,
|
| 333 |
-
"step": 1125
|
| 334 |
-
},
|
| 335 |
-
{
|
| 336 |
-
"epoch": 0.23,
|
| 337 |
-
"grad_norm": 4.038721084594727,
|
| 338 |
-
"learning_rate": 7.751610305958132e-06,
|
| 339 |
-
"loss": 0.2316,
|
| 340 |
-
"step": 1150
|
| 341 |
-
},
|
| 342 |
-
{
|
| 343 |
-
"epoch": 0.23,
|
| 344 |
-
"grad_norm": 5.786244869232178,
|
| 345 |
-
"learning_rate": 7.701288244766507e-06,
|
| 346 |
-
"loss": 0.2238,
|
| 347 |
-
"step": 1175
|
| 348 |
-
},
|
| 349 |
-
{
|
| 350 |
-
"epoch": 0.24,
|
| 351 |
-
"grad_norm": 7.598459243774414,
|
| 352 |
-
"learning_rate": 7.65096618357488e-06,
|
| 353 |
-
"loss": 0.217,
|
| 354 |
-
"step": 1200
|
| 355 |
-
},
|
| 356 |
-
{
|
| 357 |
-
"epoch": 0.24,
|
| 358 |
-
"grad_norm": 8.268415451049805,
|
| 359 |
-
"learning_rate": 7.600644122383254e-06,
|
| 360 |
-
"loss": 0.2337,
|
| 361 |
-
"step": 1225
|
| 362 |
-
},
|
| 363 |
-
{
|
| 364 |
-
"epoch": 0.25,
|
| 365 |
-
"grad_norm": 7.092071533203125,
|
| 366 |
-
"learning_rate": 7.5503220611916275e-06,
|
| 367 |
-
"loss": 0.2106,
|
| 368 |
-
"step": 1250
|
| 369 |
-
},
|
| 370 |
-
{
|
| 371 |
-
"epoch": 0.26,
|
| 372 |
-
"grad_norm": 5.1447954177856445,
|
| 373 |
-
"learning_rate": 7.500000000000001e-06,
|
| 374 |
-
"loss": 0.2088,
|
| 375 |
-
"step": 1275
|
| 376 |
-
},
|
| 377 |
-
{
|
| 378 |
-
"epoch": 0.26,
|
| 379 |
-
"grad_norm": 6.847914695739746,
|
| 380 |
-
"learning_rate": 7.449677938808374e-06,
|
| 381 |
-
"loss": 0.2151,
|
| 382 |
-
"step": 1300
|
| 383 |
-
},
|
| 384 |
-
{
|
| 385 |
-
"epoch": 0.27,
|
| 386 |
-
"grad_norm": 5.531068801879883,
|
| 387 |
-
"learning_rate": 7.399355877616747e-06,
|
| 388 |
-
"loss": 0.2031,
|
| 389 |
-
"step": 1325
|
| 390 |
-
},
|
| 391 |
-
{
|
| 392 |
-
"epoch": 0.27,
|
| 393 |
-
"grad_norm": 5.454577445983887,
|
| 394 |
-
"learning_rate": 7.349033816425122e-06,
|
| 395 |
-
"loss": 0.1933,
|
| 396 |
-
"step": 1350
|
| 397 |
-
},
|
| 398 |
-
{
|
| 399 |
-
"epoch": 0.28,
|
| 400 |
-
"grad_norm": 5.625954627990723,
|
| 401 |
-
"learning_rate": 7.298711755233495e-06,
|
| 402 |
-
"loss": 0.2437,
|
| 403 |
-
"step": 1375
|
| 404 |
-
},
|
| 405 |
-
{
|
| 406 |
-
"epoch": 0.28,
|
| 407 |
-
"grad_norm": 5.812635898590088,
|
| 408 |
-
"learning_rate": 7.248389694041869e-06,
|
| 409 |
-
"loss": 0.2787,
|
| 410 |
-
"step": 1400
|
| 411 |
-
},
|
| 412 |
-
{
|
| 413 |
-
"epoch": 0.28,
|
| 414 |
-
"grad_norm": 6.263451099395752,
|
| 415 |
-
"learning_rate": 7.1980676328502416e-06,
|
| 416 |
-
"loss": 0.3031,
|
| 417 |
-
"step": 1425
|
| 418 |
-
},
|
| 419 |
-
{
|
| 420 |
-
"epoch": 0.29,
|
| 421 |
-
"grad_norm": 5.698704242706299,
|
| 422 |
-
"learning_rate": 7.147745571658615e-06,
|
| 423 |
-
"loss": 0.2982,
|
| 424 |
-
"step": 1450
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch": 0.
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
-
"step":
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch": 0.
|
| 435 |
-
"grad_norm":
|
| 436 |
-
"learning_rate":
|
| 437 |
-
"loss": 0.
|
| 438 |
-
"step":
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 0.3,
|
| 442 |
-
"grad_norm":
|
| 443 |
-
"learning_rate":
|
| 444 |
-
"loss": 0.
|
| 445 |
-
"step":
|
| 446 |
-
},
|
| 447 |
-
{
|
| 448 |
-
"epoch": 0.31,
|
| 449 |
-
"grad_norm": 8.562033653259277,
|
| 450 |
-
"learning_rate": 6.94645732689211e-06,
|
| 451 |
-
"loss": 0.2733,
|
| 452 |
-
"step": 1550
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 0.32,
|
| 456 |
-
"grad_norm": 6.5205888748168945,
|
| 457 |
-
"learning_rate": 6.896135265700483e-06,
|
| 458 |
-
"loss": 0.2294,
|
| 459 |
-
"step": 1575
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 0.32,
|
| 463 |
-
"grad_norm": 5.271300792694092,
|
| 464 |
-
"learning_rate": 6.845813204508857e-06,
|
| 465 |
-
"loss": 0.2271,
|
| 466 |
-
"step": 1600
|
| 467 |
-
},
|
| 468 |
-
{
|
| 469 |
-
"epoch": 0.33,
|
| 470 |
-
"grad_norm": 9.590472221374512,
|
| 471 |
-
"learning_rate": 6.795491143317231e-06,
|
| 472 |
-
"loss": 0.2366,
|
| 473 |
-
"step": 1625
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"epoch": 0.33,
|
| 477 |
-
"grad_norm": 5.548666000366211,
|
| 478 |
-
"learning_rate": 6.7451690821256045e-06,
|
| 479 |
-
"loss": 0.188,
|
| 480 |
-
"step": 1650
|
| 481 |
-
},
|
| 482 |
-
{
|
| 483 |
-
"epoch": 0.34,
|
| 484 |
-
"grad_norm": 7.139528751373291,
|
| 485 |
-
"learning_rate": 6.694847020933978e-06,
|
| 486 |
-
"loss": 0.1934,
|
| 487 |
-
"step": 1675
|
| 488 |
-
},
|
| 489 |
-
{
|
| 490 |
-
"epoch": 0.34,
|
| 491 |
-
"grad_norm": 7.057426929473877,
|
| 492 |
-
"learning_rate": 6.644524959742351e-06,
|
| 493 |
-
"loss": 0.2008,
|
| 494 |
-
"step": 1700
|
| 495 |
-
},
|
| 496 |
-
{
|
| 497 |
-
"epoch": 0.34,
|
| 498 |
-
"grad_norm": 5.961209774017334,
|
| 499 |
-
"learning_rate": 6.594202898550725e-06,
|
| 500 |
-
"loss": 0.2003,
|
| 501 |
-
"step": 1725
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 0.35,
|
| 505 |
-
"grad_norm": 6.465278625488281,
|
| 506 |
-
"learning_rate": 6.543880837359099e-06,
|
| 507 |
-
"loss": 0.1979,
|
| 508 |
-
"step": 1750
|
| 509 |
-
},
|
| 510 |
-
{
|
| 511 |
-
"epoch": 0.35,
|
| 512 |
-
"grad_norm": 4.620067119598389,
|
| 513 |
-
"learning_rate": 6.493558776167472e-06,
|
| 514 |
-
"loss": 0.2111,
|
| 515 |
-
"step": 1775
|
| 516 |
-
},
|
| 517 |
-
{
|
| 518 |
-
"epoch": 0.36,
|
| 519 |
-
"grad_norm": 6.613126754760742,
|
| 520 |
-
"learning_rate": 6.443236714975846e-06,
|
| 521 |
-
"loss": 0.2045,
|
| 522 |
-
"step": 1800
|
| 523 |
-
},
|
| 524 |
-
{
|
| 525 |
-
"epoch": 0.36,
|
| 526 |
-
"grad_norm": 4.759819507598877,
|
| 527 |
-
"learning_rate": 6.3929146537842194e-06,
|
| 528 |
-
"loss": 0.1908,
|
| 529 |
-
"step": 1825
|
| 530 |
-
},
|
| 531 |
-
{
|
| 532 |
-
"epoch": 0.37,
|
| 533 |
-
"grad_norm": 7.598659515380859,
|
| 534 |
-
"learning_rate": 6.342592592592594e-06,
|
| 535 |
-
"loss": 0.2148,
|
| 536 |
-
"step": 1850
|
| 537 |
-
},
|
| 538 |
-
{
|
| 539 |
-
"epoch": 0.38,
|
| 540 |
-
"grad_norm": 7.476439952850342,
|
| 541 |
-
"learning_rate": 6.2922705314009666e-06,
|
| 542 |
-
"loss": 0.2213,
|
| 543 |
-
"step": 1875
|
| 544 |
-
},
|
| 545 |
-
{
|
| 546 |
-
"epoch": 0.38,
|
| 547 |
-
"grad_norm": 5.949714183807373,
|
| 548 |
-
"learning_rate": 6.24194847020934e-06,
|
| 549 |
-
"loss": 0.2106,
|
| 550 |
-
"step": 1900
|
| 551 |
-
},
|
| 552 |
-
{
|
| 553 |
-
"epoch": 0.39,
|
| 554 |
-
"grad_norm": 4.579805374145508,
|
| 555 |
-
"learning_rate": 6.191626409017714e-06,
|
| 556 |
-
"loss": 0.2128,
|
| 557 |
-
"step": 1925
|
| 558 |
},
|
| 559 |
{
|
| 560 |
-
"epoch": 0.
|
| 561 |
-
"grad_norm":
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"loss": 0.
|
| 564 |
-
"step":
|
| 565 |
-
},
|
| 566 |
-
{
|
| 567 |
-
"epoch": 0.4,
|
| 568 |
-
"grad_norm": 4.678502559661865,
|
| 569 |
-
"learning_rate": 6.090982286634462e-06,
|
| 570 |
-
"loss": 0.2089,
|
| 571 |
-
"step": 1975
|
| 572 |
},
|
| 573 |
{
|
| 574 |
-
"epoch": 0.
|
| 575 |
-
"grad_norm":
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
-
"step":
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 0.4,
|
| 582 |
-
"
|
| 583 |
-
"
|
| 584 |
-
"
|
| 585 |
-
"
|
| 586 |
-
"eval_wer": 48.40896036760483,
|
| 587 |
-
"step": 2000
|
| 588 |
-
},
|
| 589 |
-
{
|
| 590 |
-
"epoch": 0.41,
|
| 591 |
-
"grad_norm": 8.851235389709473,
|
| 592 |
-
"learning_rate": 5.990338164251208e-06,
|
| 593 |
-
"loss": 0.1833,
|
| 594 |
-
"step": 2025
|
| 595 |
-
},
|
| 596 |
-
{
|
| 597 |
-
"epoch": 0.41,
|
| 598 |
-
"grad_norm": 7.610278606414795,
|
| 599 |
-
"learning_rate": 5.9400161030595815e-06,
|
| 600 |
-
"loss": 0.2345,
|
| 601 |
-
"step": 2050
|
| 602 |
},
|
| 603 |
{
|
| 604 |
-
"epoch": 0.
|
| 605 |
-
"grad_norm":
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"loss": 0.
|
| 608 |
-
"step":
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch": 0.
|
| 612 |
-
"grad_norm":
|
| 613 |
-
"learning_rate":
|
| 614 |
-
"loss": 0.
|
| 615 |
-
"step":
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch": 0.
|
| 619 |
-
"grad_norm":
|
| 620 |
-
"learning_rate":
|
| 621 |
-
"loss": 0.
|
| 622 |
-
"step":
|
| 623 |
-
},
|
| 624 |
-
{
|
| 625 |
-
"epoch": 0.43,
|
| 626 |
-
"grad_norm": 5.895429611206055,
|
| 627 |
-
"learning_rate": 5.738727858293076e-06,
|
| 628 |
-
"loss": 0.2227,
|
| 629 |
-
"step": 2150
|
| 630 |
-
},
|
| 631 |
-
{
|
| 632 |
-
"epoch": 0.43,
|
| 633 |
-
"grad_norm": 6.225393295288086,
|
| 634 |
-
"learning_rate": 5.688405797101449e-06,
|
| 635 |
-
"loss": 0.2254,
|
| 636 |
-
"step": 2175
|
| 637 |
-
},
|
| 638 |
-
{
|
| 639 |
-
"epoch": 0.44,
|
| 640 |
-
"grad_norm": 4.9755401611328125,
|
| 641 |
-
"learning_rate": 5.638083735909823e-06,
|
| 642 |
-
"loss": 0.2268,
|
| 643 |
-
"step": 2200
|
| 644 |
-
},
|
| 645 |
-
{
|
| 646 |
-
"epoch": 0.45,
|
| 647 |
-
"grad_norm": 4.857475757598877,
|
| 648 |
-
"learning_rate": 5.587761674718197e-06,
|
| 649 |
-
"loss": 0.2554,
|
| 650 |
-
"step": 2225
|
| 651 |
-
},
|
| 652 |
-
{
|
| 653 |
-
"epoch": 0.45,
|
| 654 |
-
"grad_norm": 6.710099697113037,
|
| 655 |
-
"learning_rate": 5.537439613526571e-06,
|
| 656 |
-
"loss": 0.2674,
|
| 657 |
-
"step": 2250
|
| 658 |
-
},
|
| 659 |
-
{
|
| 660 |
-
"epoch": 0.46,
|
| 661 |
-
"grad_norm": 6.126322269439697,
|
| 662 |
-
"learning_rate": 5.4871175523349444e-06,
|
| 663 |
-
"loss": 0.2679,
|
| 664 |
-
"step": 2275
|
| 665 |
-
},
|
| 666 |
-
{
|
| 667 |
-
"epoch": 0.46,
|
| 668 |
-
"grad_norm": 8.071053504943848,
|
| 669 |
-
"learning_rate": 5.436795491143317e-06,
|
| 670 |
-
"loss": 0.2781,
|
| 671 |
-
"step": 2300
|
| 672 |
-
},
|
| 673 |
-
{
|
| 674 |
-
"epoch": 0.47,
|
| 675 |
-
"grad_norm": 6.979907035827637,
|
| 676 |
-
"learning_rate": 5.386473429951691e-06,
|
| 677 |
-
"loss": 0.2821,
|
| 678 |
-
"step": 2325
|
| 679 |
-
},
|
| 680 |
-
{
|
| 681 |
-
"epoch": 0.47,
|
| 682 |
-
"grad_norm": 7.387816429138184,
|
| 683 |
-
"learning_rate": 5.336151368760065e-06,
|
| 684 |
-
"loss": 0.3021,
|
| 685 |
-
"step": 2350
|
| 686 |
-
},
|
| 687 |
-
{
|
| 688 |
-
"epoch": 0.47,
|
| 689 |
-
"grad_norm": 4.772555351257324,
|
| 690 |
-
"learning_rate": 5.285829307568439e-06,
|
| 691 |
-
"loss": 0.2713,
|
| 692 |
-
"step": 2375
|
| 693 |
-
},
|
| 694 |
-
{
|
| 695 |
-
"epoch": 0.48,
|
| 696 |
-
"grad_norm": 6.187594890594482,
|
| 697 |
-
"learning_rate": 5.235507246376812e-06,
|
| 698 |
-
"loss": 0.2674,
|
| 699 |
-
"step": 2400
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 0.
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate":
|
| 705 |
-
"loss": 0.
|
| 706 |
-
"step":
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch": 0.
|
| 710 |
-
"grad_norm":
|
| 711 |
-
"learning_rate":
|
| 712 |
-
"loss": 0.
|
| 713 |
-
"step":
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch": 0.
|
| 717 |
-
"grad_norm":
|
| 718 |
-
"learning_rate":
|
| 719 |
-
"loss": 0.
|
| 720 |
-
"step":
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch": 0.
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss": 0.
|
| 727 |
-
"step":
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch": 0.
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
-
"step":
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch": 0.
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"loss": 0.
|
| 741 |
-
"step":
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch": 0.
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate":
|
| 747 |
-
"loss": 0.
|
| 748 |
-
"step":
|
| 749 |
},
|
| 750 |
{
|
| 751 |
-
"epoch": 0.
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate":
|
| 754 |
-
"loss": 0.
|
| 755 |
-
"step":
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch": 0.
|
| 759 |
-
"grad_norm": 4.
|
| 760 |
-
"learning_rate":
|
| 761 |
-
"loss": 0.
|
| 762 |
-
"step":
|
| 763 |
},
|
| 764 |
{
|
| 765 |
-
"epoch": 0.
|
| 766 |
-
"grad_norm":
|
| 767 |
-
"learning_rate":
|
| 768 |
-
"loss": 0.
|
| 769 |
-
"step":
|
| 770 |
},
|
| 771 |
{
|
| 772 |
-
"epoch": 0.
|
| 773 |
-
"grad_norm":
|
| 774 |
-
"learning_rate":
|
| 775 |
-
"loss": 0.
|
| 776 |
-
"step":
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"epoch": 0.
|
| 780 |
-
"grad_norm":
|
| 781 |
-
"learning_rate":
|
| 782 |
-
"loss": 0.
|
| 783 |
-
"step":
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"epoch": 0.
|
| 787 |
-
"grad_norm":
|
| 788 |
-
"learning_rate":
|
| 789 |
-
"loss": 0.
|
| 790 |
-
"step":
|
| 791 |
},
|
| 792 |
{
|
| 793 |
-
"epoch": 0.
|
| 794 |
-
"grad_norm":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
-
"step":
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch": 0
|
| 801 |
-
"grad_norm":
|
| 802 |
-
"learning_rate":
|
| 803 |
-
"loss": 0.
|
| 804 |
-
"step":
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch":
|
| 808 |
-
"grad_norm":
|
| 809 |
-
"learning_rate":
|
| 810 |
-
"loss": 0.
|
| 811 |
-
"step":
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"epoch":
|
| 815 |
-
"grad_norm":
|
| 816 |
-
"learning_rate":
|
| 817 |
-
"loss": 0.
|
| 818 |
-
"step":
|
| 819 |
},
|
| 820 |
{
|
| 821 |
-
"epoch":
|
| 822 |
-
"grad_norm":
|
| 823 |
-
"learning_rate":
|
| 824 |
-
"loss": 0.
|
| 825 |
-
"step":
|
| 826 |
},
|
| 827 |
{
|
| 828 |
-
"epoch":
|
| 829 |
-
"grad_norm":
|
| 830 |
-
"learning_rate":
|
| 831 |
-
"loss": 0.
|
| 832 |
-
"step":
|
| 833 |
},
|
| 834 |
{
|
| 835 |
-
"epoch":
|
| 836 |
-
"grad_norm":
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 0.
|
| 839 |
-
"step":
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"grad_norm":
|
| 844 |
-
"learning_rate":
|
| 845 |
-
"loss": 0.
|
| 846 |
-
"step":
|
| 847 |
},
|
| 848 |
{
|
| 849 |
-
"epoch":
|
| 850 |
-
"grad_norm":
|
| 851 |
-
"learning_rate":
|
| 852 |
-
"loss": 0.
|
| 853 |
-
"step":
|
| 854 |
},
|
| 855 |
{
|
| 856 |
-
"epoch":
|
| 857 |
-
"grad_norm":
|
| 858 |
-
"learning_rate":
|
| 859 |
-
"loss": 0.
|
| 860 |
-
"step":
|
| 861 |
},
|
| 862 |
{
|
| 863 |
-
"epoch":
|
| 864 |
-
"grad_norm":
|
| 865 |
-
"learning_rate":
|
| 866 |
-
"loss": 0.
|
| 867 |
-
"step":
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"
|
| 872 |
-
"
|
| 873 |
-
"
|
| 874 |
-
"
|
| 875 |
-
"eval_wer": 42.12521539345204,
|
| 876 |
-
"step": 3000
|
| 877 |
-
},
|
| 878 |
-
{
|
| 879 |
-
"epoch": 0.6,
|
| 880 |
-
"grad_norm": 5.306319713592529,
|
| 881 |
-
"learning_rate": 3.979468599033817e-06,
|
| 882 |
-
"loss": 0.1559,
|
| 883 |
-
"step": 3025
|
| 884 |
},
|
| 885 |
{
|
| 886 |
-
"epoch":
|
| 887 |
-
"grad_norm":
|
| 888 |
-
"learning_rate":
|
| 889 |
-
"loss": 0.
|
| 890 |
-
"step":
|
| 891 |
},
|
| 892 |
{
|
| 893 |
-
"epoch":
|
| 894 |
-
"grad_norm":
|
| 895 |
-
"learning_rate":
|
| 896 |
-
"loss": 0.
|
| 897 |
-
"step":
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"epoch":
|
| 901 |
-
"grad_norm":
|
| 902 |
-
"learning_rate":
|
| 903 |
-
"loss": 0.
|
| 904 |
-
"step":
|
| 905 |
},
|
| 906 |
{
|
| 907 |
-
"epoch":
|
| 908 |
-
"grad_norm":
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss": 0.
|
| 911 |
-
"step":
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch":
|
| 915 |
-
"grad_norm":
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"loss": 0.
|
| 918 |
-
"step":
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch":
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate":
|
| 924 |
-
"loss": 0.
|
| 925 |
-
"step":
|
| 926 |
},
|
| 927 |
{
|
| 928 |
-
"epoch":
|
| 929 |
-
"grad_norm":
|
| 930 |
-
"learning_rate":
|
| 931 |
-
"loss": 0.
|
| 932 |
-
"step":
|
| 933 |
},
|
| 934 |
{
|
| 935 |
-
"epoch":
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate":
|
| 938 |
-
"loss": 0.
|
| 939 |
-
"step":
|
| 940 |
},
|
| 941 |
{
|
| 942 |
-
"epoch":
|
| 943 |
-
"grad_norm":
|
| 944 |
-
"learning_rate":
|
| 945 |
-
"loss": 0.
|
| 946 |
-
"step":
|
| 947 |
},
|
| 948 |
{
|
| 949 |
-
"epoch":
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
-
"step":
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch":
|
| 957 |
-
"grad_norm":
|
| 958 |
-
"learning_rate":
|
| 959 |
-
"loss": 0.
|
| 960 |
-
"step":
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch":
|
| 964 |
-
"grad_norm":
|
| 965 |
-
"learning_rate":
|
| 966 |
-
"loss": 0.
|
| 967 |
-
"step":
|
| 968 |
},
|
| 969 |
{
|
| 970 |
-
"epoch":
|
| 971 |
-
"grad_norm":
|
| 972 |
-
"learning_rate":
|
| 973 |
-
"loss": 0.
|
| 974 |
-
"step":
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"grad_norm":
|
| 979 |
-
"learning_rate":
|
| 980 |
-
"loss": 0.
|
| 981 |
-
"step":
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"epoch":
|
| 985 |
-
"grad_norm":
|
| 986 |
-
"learning_rate":
|
| 987 |
-
"loss": 0.
|
| 988 |
-
"step":
|
| 989 |
},
|
| 990 |
{
|
| 991 |
-
"epoch":
|
| 992 |
-
"grad_norm":
|
| 993 |
-
"learning_rate":
|
| 994 |
-
"loss": 0.
|
| 995 |
-
"step":
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch":
|
| 999 |
-
"grad_norm":
|
| 1000 |
-
"learning_rate":
|
| 1001 |
-
"loss": 0.
|
| 1002 |
-
"step":
|
| 1003 |
},
|
| 1004 |
{
|
| 1005 |
-
"epoch":
|
| 1006 |
-
"grad_norm":
|
| 1007 |
-
"learning_rate":
|
| 1008 |
-
"loss": 0.
|
| 1009 |
-
"step":
|
| 1010 |
},
|
| 1011 |
{
|
| 1012 |
-
"epoch":
|
| 1013 |
-
"grad_norm":
|
| 1014 |
-
"learning_rate":
|
| 1015 |
-
"loss": 0.
|
| 1016 |
-
"step":
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch": 0
|
| 1020 |
-
"grad_norm":
|
| 1021 |
-
"learning_rate":
|
| 1022 |
-
"loss": 0.
|
| 1023 |
-
"step":
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
-
"grad_norm":
|
| 1028 |
-
"learning_rate":
|
| 1029 |
-
"loss": 0.
|
| 1030 |
-
"step":
|
| 1031 |
},
|
| 1032 |
{
|
| 1033 |
-
"epoch":
|
| 1034 |
-
"grad_norm":
|
| 1035 |
-
"learning_rate":
|
| 1036 |
-
"loss": 0.
|
| 1037 |
-
"step":
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
-
"epoch":
|
| 1041 |
-
"grad_norm":
|
| 1042 |
-
"learning_rate":
|
| 1043 |
-
"loss": 0.
|
| 1044 |
-
"step":
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
-
"epoch":
|
| 1048 |
-
"grad_norm":
|
| 1049 |
-
"learning_rate":
|
| 1050 |
-
"loss": 0.
|
| 1051 |
-
"step":
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
-
"epoch":
|
| 1055 |
-
"grad_norm":
|
| 1056 |
-
"learning_rate":
|
| 1057 |
-
"loss": 0.
|
| 1058 |
-
"step":
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
-
"epoch":
|
| 1062 |
-
"grad_norm":
|
| 1063 |
-
"learning_rate":
|
| 1064 |
-
"loss": 0.
|
| 1065 |
-
"step":
|
| 1066 |
},
|
| 1067 |
{
|
| 1068 |
-
"epoch":
|
| 1069 |
-
"grad_norm":
|
| 1070 |
-
"learning_rate":
|
| 1071 |
-
"loss": 0.
|
| 1072 |
-
"step":
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
-
"epoch":
|
| 1076 |
-
"grad_norm":
|
| 1077 |
-
"learning_rate":
|
| 1078 |
-
"loss": 0.
|
| 1079 |
-
"step":
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
-
"epoch":
|
| 1083 |
-
"grad_norm":
|
| 1084 |
-
"learning_rate":
|
| 1085 |
-
"loss": 0.
|
| 1086 |
-
"step":
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
-
"epoch":
|
| 1090 |
-
"grad_norm":
|
| 1091 |
-
"learning_rate":
|
| 1092 |
-
"loss": 0.
|
| 1093 |
-
"step":
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
-
"epoch":
|
| 1097 |
-
"grad_norm":
|
| 1098 |
-
"learning_rate":
|
| 1099 |
-
"loss": 0.
|
| 1100 |
-
"step":
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
-
"epoch":
|
| 1104 |
-
"grad_norm":
|
| 1105 |
-
"learning_rate":
|
| 1106 |
-
"loss": 0.
|
| 1107 |
-
"step":
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
-
"epoch":
|
| 1111 |
-
"grad_norm":
|
| 1112 |
-
"learning_rate":
|
| 1113 |
-
"loss": 0.
|
| 1114 |
-
"step":
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
-
"epoch":
|
| 1118 |
-
"grad_norm":
|
| 1119 |
-
"learning_rate":
|
| 1120 |
-
"loss": 0.
|
| 1121 |
-
"step":
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
-
"epoch":
|
| 1125 |
-
"grad_norm":
|
| 1126 |
-
"learning_rate":
|
| 1127 |
-
"loss": 0.
|
| 1128 |
-
"step":
|
| 1129 |
},
|
| 1130 |
{
|
| 1131 |
-
"epoch":
|
| 1132 |
-
"grad_norm":
|
| 1133 |
-
"learning_rate":
|
| 1134 |
-
"loss": 0.
|
| 1135 |
-
"step":
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
-
"epoch":
|
| 1139 |
-
"grad_norm":
|
| 1140 |
-
"learning_rate":
|
| 1141 |
-
"loss": 0.
|
| 1142 |
-
"step":
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
-
"epoch":
|
| 1146 |
-
"grad_norm":
|
| 1147 |
-
"learning_rate":
|
| 1148 |
-
"loss": 0.
|
| 1149 |
-
"step":
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
-
"epoch":
|
| 1153 |
-
"grad_norm":
|
| 1154 |
-
"learning_rate":
|
| 1155 |
-
"loss": 0.
|
| 1156 |
-
"step":
|
| 1157 |
},
|
| 1158 |
{
|
| 1159 |
-
"epoch":
|
| 1160 |
-
"
|
| 1161 |
-
"
|
| 1162 |
-
"
|
| 1163 |
-
"
|
| 1164 |
-
"eval_wer": 42.05628948879954,
|
| 1165 |
-
"step": 4000
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
-
"epoch":
|
| 1169 |
-
"grad_norm":
|
| 1170 |
-
"learning_rate":
|
| 1171 |
-
"loss": 0.
|
| 1172 |
-
"step":
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"epoch":
|
| 1176 |
-
"grad_norm":
|
| 1177 |
-
"learning_rate":
|
| 1178 |
-
"loss": 0.
|
| 1179 |
-
"step":
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
-
"epoch":
|
| 1183 |
-
"grad_norm":
|
| 1184 |
-
"learning_rate":
|
| 1185 |
-
"loss": 0.
|
| 1186 |
-
"step":
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
-
"epoch":
|
| 1190 |
-
"grad_norm":
|
| 1191 |
-
"learning_rate":
|
| 1192 |
-
"loss": 0.
|
| 1193 |
-
"step":
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
-
"epoch":
|
| 1197 |
-
"grad_norm":
|
| 1198 |
-
"learning_rate":
|
| 1199 |
-
"loss": 0.
|
| 1200 |
-
"step":
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch":
|
| 1204 |
-
"grad_norm":
|
| 1205 |
-
"learning_rate":
|
| 1206 |
-
"loss": 0.
|
| 1207 |
-
"step":
|
| 1208 |
},
|
| 1209 |
{
|
| 1210 |
-
"epoch":
|
| 1211 |
-
"grad_norm":
|
| 1212 |
-
"learning_rate":
|
| 1213 |
-
"loss": 0.
|
| 1214 |
-
"step":
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
-
"epoch":
|
| 1218 |
-
"grad_norm":
|
| 1219 |
-
"learning_rate":
|
| 1220 |
-
"loss": 0.
|
| 1221 |
-
"step":
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
-
"epoch":
|
| 1225 |
-
"grad_norm":
|
| 1226 |
-
"learning_rate":
|
| 1227 |
-
"loss": 0.
|
| 1228 |
-
"step":
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
-
"epoch": 0
|
| 1232 |
-
"grad_norm":
|
| 1233 |
-
"learning_rate":
|
| 1234 |
-
"loss": 0.
|
| 1235 |
-
"step":
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
-
"epoch":
|
| 1239 |
-
"grad_norm":
|
| 1240 |
-
"learning_rate":
|
| 1241 |
-
"loss": 0.
|
| 1242 |
-
"step":
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"epoch":
|
| 1246 |
-
"grad_norm":
|
| 1247 |
-
"learning_rate":
|
| 1248 |
-
"loss": 0.
|
| 1249 |
-
"step":
|
| 1250 |
},
|
| 1251 |
{
|
| 1252 |
-
"epoch":
|
| 1253 |
-
"grad_norm":
|
| 1254 |
-
"learning_rate":
|
| 1255 |
-
"loss": 0.
|
| 1256 |
-
"step":
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
-
"epoch":
|
| 1260 |
-
"grad_norm":
|
| 1261 |
-
"learning_rate":
|
| 1262 |
-
"loss": 0.
|
| 1263 |
-
"step":
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
-
"epoch":
|
| 1267 |
-
"grad_norm":
|
| 1268 |
-
"learning_rate":
|
| 1269 |
-
"loss": 0.
|
| 1270 |
-
"step":
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
-
"epoch":
|
| 1274 |
-
"grad_norm":
|
| 1275 |
-
"learning_rate":
|
| 1276 |
-
"loss": 0.
|
| 1277 |
-
"step":
|
| 1278 |
},
|
| 1279 |
{
|
| 1280 |
-
"epoch":
|
| 1281 |
-
"grad_norm":
|
| 1282 |
-
"learning_rate":
|
| 1283 |
-
"loss": 0.
|
| 1284 |
-
"step":
|
| 1285 |
},
|
| 1286 |
{
|
| 1287 |
-
"epoch":
|
| 1288 |
-
"grad_norm":
|
| 1289 |
-
"learning_rate":
|
| 1290 |
-
"loss": 0.
|
| 1291 |
-
"step":
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
-
"epoch":
|
| 1295 |
-
"grad_norm":
|
| 1296 |
-
"learning_rate":
|
| 1297 |
-
"loss": 0.
|
| 1298 |
-
"step":
|
| 1299 |
},
|
| 1300 |
{
|
| 1301 |
-
"epoch":
|
| 1302 |
-
"grad_norm":
|
| 1303 |
-
"learning_rate":
|
| 1304 |
-
"loss": 0.
|
| 1305 |
-
"step":
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
-
"epoch":
|
| 1309 |
-
"grad_norm":
|
| 1310 |
-
"learning_rate":
|
| 1311 |
-
"loss": 0.
|
| 1312 |
-
"step":
|
| 1313 |
},
|
| 1314 |
{
|
| 1315 |
-
"epoch":
|
| 1316 |
-
"grad_norm":
|
| 1317 |
-
"learning_rate":
|
| 1318 |
-
"loss": 0.
|
| 1319 |
-
"step":
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
-
"epoch":
|
| 1323 |
-
"grad_norm":
|
| 1324 |
-
"learning_rate":
|
| 1325 |
-
"loss": 0.
|
| 1326 |
-
"step":
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
-
"epoch":
|
| 1330 |
-
"grad_norm":
|
| 1331 |
-
"learning_rate":
|
| 1332 |
-
"loss": 0.
|
| 1333 |
-
"step":
|
| 1334 |
},
|
| 1335 |
{
|
| 1336 |
-
"epoch":
|
| 1337 |
-
"grad_norm":
|
| 1338 |
-
"learning_rate":
|
| 1339 |
-
"loss": 0.
|
| 1340 |
-
"step":
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
-
"epoch":
|
| 1344 |
-
"grad_norm":
|
| 1345 |
-
"learning_rate":
|
| 1346 |
-
"loss": 0.
|
| 1347 |
-
"step":
|
| 1348 |
},
|
| 1349 |
{
|
| 1350 |
-
"epoch":
|
| 1351 |
-
"grad_norm":
|
| 1352 |
-
"learning_rate":
|
| 1353 |
-
"loss": 0.
|
| 1354 |
-
"step":
|
| 1355 |
},
|
| 1356 |
{
|
| 1357 |
-
"epoch":
|
| 1358 |
-
"grad_norm":
|
| 1359 |
-
"learning_rate":
|
| 1360 |
-
"loss": 0.
|
| 1361 |
-
"step":
|
| 1362 |
},
|
| 1363 |
{
|
| 1364 |
-
"epoch":
|
| 1365 |
-
"grad_norm":
|
| 1366 |
-
"learning_rate":
|
| 1367 |
-
"loss": 0.
|
| 1368 |
-
"step":
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
-
"epoch":
|
| 1372 |
-
"grad_norm":
|
| 1373 |
-
"learning_rate":
|
| 1374 |
-
"loss": 0.
|
| 1375 |
-
"step":
|
| 1376 |
},
|
| 1377 |
{
|
| 1378 |
-
"epoch":
|
| 1379 |
-
"grad_norm":
|
| 1380 |
-
"learning_rate":
|
| 1381 |
-
"loss": 0.
|
| 1382 |
-
"step":
|
| 1383 |
},
|
| 1384 |
{
|
| 1385 |
-
"epoch":
|
| 1386 |
-
"grad_norm":
|
| 1387 |
-
"learning_rate":
|
| 1388 |
-
"loss": 0.
|
| 1389 |
-
"step":
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
-
"epoch":
|
| 1393 |
-
"grad_norm":
|
| 1394 |
-
"learning_rate": 3.
|
| 1395 |
-
"loss": 0.
|
| 1396 |
-
"step":
|
| 1397 |
},
|
| 1398 |
{
|
| 1399 |
-
"epoch":
|
| 1400 |
-
"grad_norm":
|
| 1401 |
-
"learning_rate": 3.
|
| 1402 |
-
"loss": 0.
|
| 1403 |
-
"step":
|
| 1404 |
},
|
| 1405 |
{
|
| 1406 |
-
"epoch":
|
| 1407 |
-
"grad_norm":
|
| 1408 |
-
"learning_rate":
|
| 1409 |
-
"loss": 0.
|
| 1410 |
-
"step":
|
| 1411 |
},
|
| 1412 |
{
|
| 1413 |
-
"epoch":
|
| 1414 |
-
"grad_norm":
|
| 1415 |
-
"learning_rate":
|
| 1416 |
-
"loss": 0.
|
| 1417 |
-
"step":
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
-
"epoch":
|
| 1421 |
-
"grad_norm":
|
| 1422 |
-
"learning_rate":
|
| 1423 |
-
"loss": 0.
|
| 1424 |
-
"step":
|
| 1425 |
},
|
| 1426 |
{
|
| 1427 |
-
"epoch":
|
| 1428 |
-
"grad_norm":
|
| 1429 |
-
"learning_rate":
|
| 1430 |
-
"loss": 0.
|
| 1431 |
-
"step":
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
-
"epoch":
|
| 1435 |
-
"grad_norm":
|
| 1436 |
-
"learning_rate":
|
| 1437 |
-
"loss": 0.
|
| 1438 |
-
"step":
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
-
"epoch":
|
| 1442 |
-
"grad_norm":
|
| 1443 |
-
"learning_rate":
|
| 1444 |
-
"loss": 0.
|
| 1445 |
-
"step":
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
-
"epoch":
|
| 1449 |
-
"eval_loss": 0.
|
| 1450 |
-
"eval_runtime": 287.
|
| 1451 |
-
"eval_samples_per_second": 3.
|
| 1452 |
"eval_steps_per_second": 0.435,
|
| 1453 |
-
"eval_wer":
|
| 1454 |
-
"step":
|
| 1455 |
}
|
| 1456 |
],
|
| 1457 |
-
"logging_steps":
|
| 1458 |
-
"max_steps":
|
| 1459 |
"num_input_tokens_seen": 0,
|
| 1460 |
-
"num_train_epochs":
|
| 1461 |
-
"save_steps":
|
| 1462 |
-
"
|
| 1463 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1464 |
"trial_name": null,
|
| 1465 |
"trial_params": null
|
| 1466 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 19.701321079839175,
|
| 3 |
+
"best_model_checkpoint": "./whisper-ft-2/checkpoint-600",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 600,
|
| 6 |
+
"global_step": 600,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03333333333333333,
|
| 13 |
+
"grad_norm": 1.419047236442566,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.0144,
|
| 16 |
+
"step": 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06666666666666667,
|
| 20 |
+
"grad_norm": 2.3382816314697266,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 0.0143,
|
| 23 |
+
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 0.8841551542282104,
|
| 28 |
+
"learning_rate": 9.943820224719102e-06,
|
| 29 |
+
"loss": 0.0126,
|
| 30 |
+
"step": 15
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.13333333333333333,
|
| 34 |
+
"grad_norm": 2.564495325088501,
|
| 35 |
+
"learning_rate": 9.887640449438202e-06,
|
| 36 |
+
"loss": 0.0144,
|
| 37 |
+
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.16666666666666666,
|
| 41 |
+
"grad_norm": 1.858008623123169,
|
| 42 |
+
"learning_rate": 9.831460674157303e-06,
|
| 43 |
+
"loss": 0.0097,
|
| 44 |
+
"step": 25
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.2,
|
| 48 |
+
"grad_norm": 1.3031666278839111,
|
| 49 |
+
"learning_rate": 9.775280898876405e-06,
|
| 50 |
+
"loss": 0.0129,
|
| 51 |
+
"step": 30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.23333333333333334,
|
| 55 |
+
"grad_norm": 3.633366584777832,
|
| 56 |
+
"learning_rate": 9.719101123595506e-06,
|
| 57 |
+
"loss": 0.0083,
|
| 58 |
+
"step": 35
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.26666666666666666,
|
| 62 |
+
"grad_norm": 1.9989734888076782,
|
| 63 |
+
"learning_rate": 9.662921348314608e-06,
|
| 64 |
+
"loss": 0.0122,
|
| 65 |
+
"step": 40
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.3,
|
| 69 |
+
"grad_norm": 0.9710230827331543,
|
| 70 |
+
"learning_rate": 9.60674157303371e-06,
|
| 71 |
+
"loss": 0.007,
|
| 72 |
+
"step": 45
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.3333333333333333,
|
| 76 |
+
"grad_norm": 1.8736152648925781,
|
| 77 |
+
"learning_rate": 9.55056179775281e-06,
|
| 78 |
+
"loss": 0.0122,
|
| 79 |
+
"step": 50
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.36666666666666664,
|
| 83 |
+
"grad_norm": 0.9838809370994568,
|
| 84 |
+
"learning_rate": 9.49438202247191e-06,
|
| 85 |
+
"loss": 0.005,
|
| 86 |
+
"step": 55
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.4,
|
| 90 |
+
"grad_norm": 2.429737091064453,
|
| 91 |
+
"learning_rate": 9.438202247191012e-06,
|
| 92 |
+
"loss": 0.0115,
|
| 93 |
+
"step": 60
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.43333333333333335,
|
| 97 |
+
"grad_norm": 1.108884572982788,
|
| 98 |
+
"learning_rate": 9.382022471910113e-06,
|
| 99 |
+
"loss": 0.0058,
|
| 100 |
+
"step": 65
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.4666666666666667,
|
| 104 |
+
"grad_norm": 1.1713858842849731,
|
| 105 |
+
"learning_rate": 9.325842696629213e-06,
|
| 106 |
+
"loss": 0.0058,
|
| 107 |
+
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.5,
|
| 111 |
+
"grad_norm": 3.6589481830596924,
|
| 112 |
+
"learning_rate": 9.269662921348316e-06,
|
| 113 |
+
"loss": 0.0094,
|
| 114 |
+
"step": 75
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.5333333333333333,
|
| 118 |
+
"grad_norm": 1.9582005739212036,
|
| 119 |
+
"learning_rate": 9.213483146067417e-06,
|
| 120 |
+
"loss": 0.0076,
|
| 121 |
+
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.5666666666666667,
|
| 125 |
+
"grad_norm": 1.3726775646209717,
|
| 126 |
+
"learning_rate": 9.157303370786517e-06,
|
| 127 |
+
"loss": 0.0054,
|
| 128 |
+
"step": 85
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.6,
|
| 132 |
+
"grad_norm": 1.7600703239440918,
|
| 133 |
+
"learning_rate": 9.101123595505619e-06,
|
| 134 |
+
"loss": 0.0099,
|
| 135 |
+
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.6333333333333333,
|
| 139 |
+
"grad_norm": 0.9456138610839844,
|
| 140 |
+
"learning_rate": 9.04494382022472e-06,
|
| 141 |
+
"loss": 0.0047,
|
| 142 |
+
"step": 95
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.6666666666666666,
|
| 146 |
+
"grad_norm": 0.9388735294342041,
|
| 147 |
+
"learning_rate": 8.988764044943822e-06,
|
| 148 |
+
"loss": 0.0089,
|
| 149 |
+
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.7,
|
| 153 |
+
"grad_norm": 2.800154685974121,
|
| 154 |
+
"learning_rate": 8.932584269662921e-06,
|
| 155 |
+
"loss": 0.0074,
|
| 156 |
+
"step": 105
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.7333333333333333,
|
| 160 |
+
"grad_norm": 4.255383491516113,
|
| 161 |
+
"learning_rate": 8.876404494382023e-06,
|
| 162 |
+
"loss": 0.0057,
|
| 163 |
+
"step": 110
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.7666666666666667,
|
| 167 |
+
"grad_norm": 1.3430100679397583,
|
| 168 |
+
"learning_rate": 8.820224719101124e-06,
|
| 169 |
+
"loss": 0.0042,
|
| 170 |
+
"step": 115
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.8,
|
| 174 |
+
"grad_norm": 4.184587478637695,
|
| 175 |
+
"learning_rate": 8.764044943820226e-06,
|
| 176 |
+
"loss": 0.0063,
|
| 177 |
+
"step": 120
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.8333333333333334,
|
| 181 |
+
"grad_norm": 1.3114601373672485,
|
| 182 |
+
"learning_rate": 8.707865168539327e-06,
|
| 183 |
+
"loss": 0.0085,
|
| 184 |
+
"step": 125
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.8666666666666667,
|
| 188 |
+
"grad_norm": 1.0539401769638062,
|
| 189 |
+
"learning_rate": 8.651685393258428e-06,
|
| 190 |
+
"loss": 0.0072,
|
| 191 |
+
"step": 130
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.9,
|
| 195 |
+
"grad_norm": 4.378466606140137,
|
| 196 |
+
"learning_rate": 8.595505617977528e-06,
|
| 197 |
+
"loss": 0.0049,
|
| 198 |
+
"step": 135
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.9333333333333333,
|
| 202 |
+
"grad_norm": 1.5044472217559814,
|
| 203 |
+
"learning_rate": 8.53932584269663e-06,
|
| 204 |
+
"loss": 0.0067,
|
| 205 |
+
"step": 140
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.9666666666666667,
|
| 209 |
+
"grad_norm": 2.2182376384735107,
|
| 210 |
+
"learning_rate": 8.483146067415731e-06,
|
| 211 |
+
"loss": 0.0091,
|
| 212 |
+
"step": 145
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 1.0,
|
| 216 |
+
"grad_norm": 0.24589230120182037,
|
| 217 |
+
"learning_rate": 8.426966292134832e-06,
|
| 218 |
+
"loss": 0.0062,
|
| 219 |
+
"step": 150
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 1.0333333333333334,
|
| 223 |
+
"grad_norm": 0.48602986335754395,
|
| 224 |
+
"learning_rate": 8.370786516853934e-06,
|
| 225 |
+
"loss": 0.0056,
|
| 226 |
+
"step": 155
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 1.0666666666666667,
|
| 230 |
+
"grad_norm": 0.2779291570186615,
|
| 231 |
+
"learning_rate": 8.314606741573035e-06,
|
| 232 |
+
"loss": 0.0031,
|
| 233 |
+
"step": 160
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 1.1,
|
| 237 |
+
"grad_norm": 2.2703235149383545,
|
| 238 |
+
"learning_rate": 8.258426966292135e-06,
|
| 239 |
+
"loss": 0.0069,
|
| 240 |
+
"step": 165
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 1.1333333333333333,
|
| 244 |
+
"grad_norm": 1.3132015466690063,
|
| 245 |
+
"learning_rate": 8.202247191011237e-06,
|
| 246 |
+
"loss": 0.006,
|
| 247 |
+
"step": 170
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 1.1666666666666667,
|
| 251 |
+
"grad_norm": 0.24997250735759735,
|
| 252 |
+
"learning_rate": 8.146067415730338e-06,
|
| 253 |
+
"loss": 0.0027,
|
| 254 |
+
"step": 175
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 1.2,
|
| 258 |
+
"grad_norm": 0.14482256770133972,
|
| 259 |
+
"learning_rate": 8.08988764044944e-06,
|
| 260 |
+
"loss": 0.0023,
|
| 261 |
+
"step": 180
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 1.2333333333333334,
|
| 265 |
+
"grad_norm": 0.7420951128005981,
|
| 266 |
+
"learning_rate": 8.033707865168539e-06,
|
| 267 |
+
"loss": 0.0064,
|
| 268 |
+
"step": 185
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 1.2666666666666666,
|
| 272 |
+
"grad_norm": 0.21342052519321442,
|
| 273 |
+
"learning_rate": 7.97752808988764e-06,
|
| 274 |
+
"loss": 0.0026,
|
| 275 |
+
"step": 190
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 1.3,
|
| 279 |
+
"grad_norm": 0.2786453664302826,
|
| 280 |
+
"learning_rate": 7.921348314606742e-06,
|
| 281 |
+
"loss": 0.0036,
|
| 282 |
+
"step": 195
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 1.3333333333333333,
|
| 286 |
+
"grad_norm": 0.36970847845077515,
|
| 287 |
+
"learning_rate": 7.865168539325843e-06,
|
| 288 |
+
"loss": 0.0046,
|
| 289 |
+
"step": 200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 1.3666666666666667,
|
| 293 |
+
"grad_norm": 0.17185050249099731,
|
| 294 |
+
"learning_rate": 7.808988764044945e-06,
|
| 295 |
+
"loss": 0.0022,
|
| 296 |
+
"step": 205
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 1.4,
|
| 300 |
+
"grad_norm": 1.4263725280761719,
|
| 301 |
+
"learning_rate": 7.752808988764046e-06,
|
| 302 |
+
"loss": 0.0034,
|
| 303 |
+
"step": 210
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 1.4333333333333333,
|
| 307 |
+
"grad_norm": 0.5556924939155579,
|
| 308 |
+
"learning_rate": 7.696629213483146e-06,
|
| 309 |
+
"loss": 0.005,
|
| 310 |
+
"step": 215
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 1.4666666666666668,
|
| 314 |
+
"grad_norm": 1.5778443813323975,
|
| 315 |
+
"learning_rate": 7.640449438202247e-06,
|
| 316 |
+
"loss": 0.0048,
|
| 317 |
+
"step": 220
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 1.5,
|
| 321 |
+
"grad_norm": 0.39948800206184387,
|
| 322 |
+
"learning_rate": 7.584269662921349e-06,
|
| 323 |
+
"loss": 0.0056,
|
| 324 |
+
"step": 225
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 1.5333333333333332,
|
| 328 |
+
"grad_norm": 1.9470425844192505,
|
| 329 |
+
"learning_rate": 7.5280898876404495e-06,
|
| 330 |
+
"loss": 0.0052,
|
| 331 |
+
"step": 230
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 1.5666666666666667,
|
| 335 |
+
"grad_norm": 1.3367302417755127,
|
| 336 |
+
"learning_rate": 7.471910112359552e-06,
|
| 337 |
+
"loss": 0.0056,
|
| 338 |
+
"step": 235
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 1.6,
|
| 342 |
+
"grad_norm": 2.2041380405426025,
|
| 343 |
+
"learning_rate": 7.415730337078652e-06,
|
| 344 |
+
"loss": 0.0026,
|
| 345 |
+
"step": 240
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 1.6333333333333333,
|
| 349 |
+
"grad_norm": 1.1062074899673462,
|
| 350 |
+
"learning_rate": 7.359550561797754e-06,
|
| 351 |
+
"loss": 0.0058,
|
| 352 |
+
"step": 245
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 1.6666666666666665,
|
| 356 |
+
"grad_norm": 1.558339238166809,
|
| 357 |
+
"learning_rate": 7.303370786516854e-06,
|
| 358 |
+
"loss": 0.0036,
|
| 359 |
+
"step": 250
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 1.7,
|
| 363 |
+
"grad_norm": 1.4909693002700806,
|
| 364 |
+
"learning_rate": 7.247191011235956e-06,
|
| 365 |
+
"loss": 0.0081,
|
| 366 |
+
"step": 255
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 1.7333333333333334,
|
| 370 |
+
"grad_norm": 1.0885131359100342,
|
| 371 |
+
"learning_rate": 7.191011235955056e-06,
|
| 372 |
+
"loss": 0.0035,
|
| 373 |
+
"step": 260
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 1.7666666666666666,
|
| 377 |
+
"grad_norm": 0.34988316893577576,
|
| 378 |
+
"learning_rate": 7.134831460674158e-06,
|
| 379 |
+
"loss": 0.0066,
|
| 380 |
+
"step": 265
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 1.8,
|
| 384 |
+
"grad_norm": 0.39315518736839294,
|
| 385 |
+
"learning_rate": 7.078651685393258e-06,
|
| 386 |
+
"loss": 0.0036,
|
| 387 |
+
"step": 270
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 1.8333333333333335,
|
| 391 |
+
"grad_norm": 0.4730512797832489,
|
| 392 |
+
"learning_rate": 7.022471910112361e-06,
|
| 393 |
+
"loss": 0.0025,
|
| 394 |
+
"step": 275
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 1.8666666666666667,
|
| 398 |
+
"grad_norm": 1.0061614513397217,
|
| 399 |
+
"learning_rate": 6.966292134831461e-06,
|
| 400 |
+
"loss": 0.0058,
|
| 401 |
+
"step": 280
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 1.9,
|
| 405 |
+
"grad_norm": 0.6261555552482605,
|
| 406 |
+
"learning_rate": 6.910112359550563e-06,
|
| 407 |
+
"loss": 0.0035,
|
| 408 |
+
"step": 285
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 1.9333333333333333,
|
| 412 |
+
"grad_norm": 0.54555344581604,
|
| 413 |
+
"learning_rate": 6.853932584269663e-06,
|
| 414 |
+
"loss": 0.0064,
|
| 415 |
+
"step": 290
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 1.9666666666666668,
|
| 419 |
+
"grad_norm": 0.26168444752693176,
|
| 420 |
+
"learning_rate": 6.797752808988765e-06,
|
| 421 |
+
"loss": 0.0063,
|
| 422 |
+
"step": 295
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 2.0,
|
| 426 |
+
"grad_norm": 2.1794660091400146,
|
| 427 |
+
"learning_rate": 6.741573033707865e-06,
|
| 428 |
+
"loss": 0.0047,
|
| 429 |
+
"step": 300
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 2.033333333333333,
|
| 433 |
+
"grad_norm": 0.07394399493932724,
|
| 434 |
+
"learning_rate": 6.685393258426967e-06,
|
| 435 |
+
"loss": 0.0033,
|
| 436 |
+
"step": 305
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 2.066666666666667,
|
| 440 |
+
"grad_norm": 0.32680198550224304,
|
| 441 |
+
"learning_rate": 6.629213483146067e-06,
|
| 442 |
+
"loss": 0.0033,
|
| 443 |
+
"step": 310
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 2.1,
|
| 447 |
+
"grad_norm": 1.1588655710220337,
|
| 448 |
+
"learning_rate": 6.57303370786517e-06,
|
| 449 |
+
"loss": 0.0044,
|
| 450 |
+
"step": 315
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 2.1333333333333333,
|
| 454 |
+
"grad_norm": 0.12549975514411926,
|
| 455 |
+
"learning_rate": 6.51685393258427e-06,
|
| 456 |
+
"loss": 0.0025,
|
| 457 |
+
"step": 320
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 2.1666666666666665,
|
| 461 |
+
"grad_norm": 1.6806613206863403,
|
| 462 |
+
"learning_rate": 6.460674157303372e-06,
|
| 463 |
+
"loss": 0.0073,
|
| 464 |
+
"step": 325
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 2.2,
|
| 468 |
+
"grad_norm": 1.2238233089447021,
|
| 469 |
+
"learning_rate": 6.404494382022472e-06,
|
| 470 |
+
"loss": 0.0063,
|
| 471 |
+
"step": 330
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 2.2333333333333334,
|
| 475 |
+
"grad_norm": 0.13195298612117767,
|
| 476 |
+
"learning_rate": 6.348314606741574e-06,
|
| 477 |
+
"loss": 0.0029,
|
| 478 |
+
"step": 335
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 2.2666666666666666,
|
| 482 |
+
"grad_norm": 0.6085399985313416,
|
| 483 |
+
"learning_rate": 6.292134831460674e-06,
|
| 484 |
+
"loss": 0.0014,
|
| 485 |
+
"step": 340
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 2.3,
|
| 489 |
+
"grad_norm": 0.1822354644536972,
|
| 490 |
+
"learning_rate": 6.235955056179776e-06,
|
| 491 |
+
"loss": 0.0017,
|
| 492 |
+
"step": 345
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 2.3333333333333335,
|
| 496 |
+
"grad_norm": 1.526226282119751,
|
| 497 |
+
"learning_rate": 6.179775280898876e-06,
|
| 498 |
+
"loss": 0.0016,
|
| 499 |
+
"step": 350
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 2.3666666666666667,
|
| 503 |
+
"grad_norm": 0.3068194091320038,
|
| 504 |
+
"learning_rate": 6.1235955056179785e-06,
|
| 505 |
+
"loss": 0.0053,
|
| 506 |
+
"step": 355
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 2.4,
|
| 510 |
+
"grad_norm": 0.3344336748123169,
|
| 511 |
+
"learning_rate": 6.06741573033708e-06,
|
| 512 |
+
"loss": 0.0034,
|
| 513 |
+
"step": 360
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 2.4333333333333336,
|
| 517 |
+
"grad_norm": 0.38717201352119446,
|
| 518 |
+
"learning_rate": 6.0112359550561805e-06,
|
| 519 |
+
"loss": 0.0032,
|
| 520 |
+
"step": 365
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 2.466666666666667,
|
| 524 |
+
"grad_norm": 3.577847957611084,
|
| 525 |
+
"learning_rate": 5.955056179775281e-06,
|
| 526 |
+
"loss": 0.0046,
|
| 527 |
+
"step": 370
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 2.5,
|
| 531 |
+
"grad_norm": 0.1945178508758545,
|
| 532 |
+
"learning_rate": 5.8988764044943826e-06,
|
| 533 |
+
"loss": 0.0016,
|
| 534 |
+
"step": 375
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 2.533333333333333,
|
| 538 |
+
"grad_norm": 0.12505494058132172,
|
| 539 |
+
"learning_rate": 5.842696629213483e-06,
|
| 540 |
+
"loss": 0.0021,
|
| 541 |
+
"step": 380
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 2.5666666666666664,
|
| 545 |
+
"grad_norm": 0.23506364226341248,
|
| 546 |
+
"learning_rate": 5.786516853932585e-06,
|
| 547 |
+
"loss": 0.0025,
|
| 548 |
+
"step": 385
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 2.6,
|
| 552 |
+
"grad_norm": 0.48507827520370483,
|
| 553 |
+
"learning_rate": 5.730337078651685e-06,
|
| 554 |
+
"loss": 0.0016,
|
| 555 |
+
"step": 390
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 2.6333333333333333,
|
| 559 |
+
"grad_norm": 0.14861613512039185,
|
| 560 |
+
"learning_rate": 5.6741573033707874e-06,
|
| 561 |
+
"loss": 0.0019,
|
| 562 |
+
"step": 395
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.6666666666666665,
|
| 566 |
+
"grad_norm": 0.07135830074548721,
|
| 567 |
+
"learning_rate": 5.617977528089889e-06,
|
| 568 |
+
"loss": 0.0018,
|
| 569 |
+
"step": 400
|
|
|
|
|
|
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 2.7,
|
| 573 |
+
"grad_norm": 0.35108867287635803,
|
| 574 |
+
"learning_rate": 5.5617977528089895e-06,
|
| 575 |
+
"loss": 0.0011,
|
| 576 |
+
"step": 405
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.7333333333333334,
|
| 580 |
+
"grad_norm": 1.0602957010269165,
|
| 581 |
+
"learning_rate": 5.50561797752809e-06,
|
| 582 |
+
"loss": 0.0015,
|
| 583 |
+
"step": 410
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 2.7666666666666666,
|
| 587 |
+
"grad_norm": 0.14372961223125458,
|
| 588 |
+
"learning_rate": 5.4494382022471915e-06,
|
| 589 |
+
"loss": 0.0023,
|
| 590 |
+
"step": 415
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.8,
|
| 594 |
+
"grad_norm": 1.5078669786453247,
|
| 595 |
+
"learning_rate": 5.393258426966292e-06,
|
| 596 |
+
"loss": 0.0017,
|
| 597 |
+
"step": 420
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 2.8333333333333335,
|
| 601 |
+
"grad_norm": 0.08180749416351318,
|
| 602 |
+
"learning_rate": 5.3370786516853935e-06,
|
| 603 |
+
"loss": 0.0022,
|
| 604 |
+
"step": 425
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.8666666666666667,
|
| 608 |
+
"grad_norm": 0.27330687642097473,
|
| 609 |
+
"learning_rate": 5.280898876404494e-06,
|
| 610 |
+
"loss": 0.0016,
|
| 611 |
+
"step": 430
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 2.9,
|
| 615 |
+
"grad_norm": 0.06831669807434082,
|
| 616 |
+
"learning_rate": 5.224719101123596e-06,
|
| 617 |
+
"loss": 0.0008,
|
| 618 |
+
"step": 435
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.9333333333333336,
|
| 622 |
+
"grad_norm": 0.3877633213996887,
|
| 623 |
+
"learning_rate": 5.168539325842698e-06,
|
| 624 |
+
"loss": 0.0022,
|
| 625 |
+
"step": 440
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 2.966666666666667,
|
| 629 |
+
"grad_norm": 2.5412757396698,
|
| 630 |
+
"learning_rate": 5.112359550561798e-06,
|
| 631 |
+
"loss": 0.003,
|
| 632 |
+
"step": 445
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 3.0,
|
| 636 |
+
"grad_norm": 0.09176287055015564,
|
| 637 |
+
"learning_rate": 5.0561797752809e-06,
|
| 638 |
+
"loss": 0.0011,
|
| 639 |
+
"step": 450
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 3.033333333333333,
|
| 643 |
+
"grad_norm": 0.03548486530780792,
|
| 644 |
+
"learning_rate": 5e-06,
|
| 645 |
+
"loss": 0.0016,
|
| 646 |
+
"step": 455
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 3.066666666666667,
|
| 650 |
+
"grad_norm": 0.11006509512662888,
|
| 651 |
+
"learning_rate": 4.943820224719101e-06,
|
| 652 |
+
"loss": 0.006,
|
| 653 |
+
"step": 460
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 3.1,
|
| 657 |
+
"grad_norm": 0.19691240787506104,
|
| 658 |
+
"learning_rate": 4.8876404494382024e-06,
|
| 659 |
+
"loss": 0.002,
|
| 660 |
+
"step": 465
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 3.1333333333333333,
|
| 664 |
+
"grad_norm": 0.30039843916893005,
|
| 665 |
+
"learning_rate": 4.831460674157304e-06,
|
| 666 |
+
"loss": 0.001,
|
| 667 |
+
"step": 470
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 3.1666666666666665,
|
| 671 |
+
"grad_norm": 0.0753551796078682,
|
| 672 |
+
"learning_rate": 4.775280898876405e-06,
|
| 673 |
+
"loss": 0.0008,
|
| 674 |
+
"step": 475
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 3.2,
|
| 678 |
+
"grad_norm": 0.12786687910556793,
|
| 679 |
+
"learning_rate": 4.719101123595506e-06,
|
| 680 |
+
"loss": 0.0026,
|
| 681 |
+
"step": 480
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 3.2333333333333334,
|
| 685 |
+
"grad_norm": 0.07095532864332199,
|
| 686 |
+
"learning_rate": 4.6629213483146065e-06,
|
| 687 |
+
"loss": 0.0004,
|
| 688 |
+
"step": 485
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 3.2666666666666666,
|
| 692 |
+
"grad_norm": 0.1347319781780243,
|
| 693 |
+
"learning_rate": 4.606741573033709e-06,
|
| 694 |
+
"loss": 0.0013,
|
| 695 |
+
"step": 490
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 3.3,
|
| 699 |
+
"grad_norm": 0.11095249652862549,
|
| 700 |
+
"learning_rate": 4.550561797752809e-06,
|
| 701 |
+
"loss": 0.0012,
|
| 702 |
+
"step": 495
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 3.3333333333333335,
|
| 706 |
+
"grad_norm": 0.10048757493495941,
|
| 707 |
+
"learning_rate": 4.494382022471911e-06,
|
| 708 |
+
"loss": 0.0014,
|
| 709 |
+
"step": 500
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 3.3666666666666667,
|
| 713 |
+
"grad_norm": 0.04148377105593681,
|
| 714 |
+
"learning_rate": 4.438202247191011e-06,
|
| 715 |
+
"loss": 0.0006,
|
| 716 |
+
"step": 505
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 3.4,
|
| 720 |
+
"grad_norm": 0.17400866746902466,
|
| 721 |
+
"learning_rate": 4.382022471910113e-06,
|
| 722 |
+
"loss": 0.0008,
|
| 723 |
+
"step": 510
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 3.4333333333333336,
|
| 727 |
+
"grad_norm": 0.2563508450984955,
|
| 728 |
+
"learning_rate": 4.325842696629214e-06,
|
| 729 |
+
"loss": 0.0027,
|
| 730 |
+
"step": 515
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 3.466666666666667,
|
| 734 |
+
"grad_norm": 0.06805615872144699,
|
| 735 |
+
"learning_rate": 4.269662921348315e-06,
|
| 736 |
+
"loss": 0.0011,
|
| 737 |
+
"step": 520
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 3.5,
|
| 741 |
+
"grad_norm": 0.20213201642036438,
|
| 742 |
+
"learning_rate": 4.213483146067416e-06,
|
| 743 |
+
"loss": 0.0031,
|
| 744 |
+
"step": 525
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 3.533333333333333,
|
| 748 |
+
"grad_norm": 0.03789810836315155,
|
| 749 |
+
"learning_rate": 4.157303370786518e-06,
|
| 750 |
+
"loss": 0.0038,
|
| 751 |
+
"step": 530
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 3.5666666666666664,
|
| 755 |
+
"grad_norm": 0.13470155000686646,
|
| 756 |
+
"learning_rate": 4.101123595505618e-06,
|
| 757 |
+
"loss": 0.0017,
|
| 758 |
+
"step": 535
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 3.6,
|
| 762 |
+
"grad_norm": 0.046678509563207626,
|
| 763 |
+
"learning_rate": 4.04494382022472e-06,
|
| 764 |
+
"loss": 0.0005,
|
| 765 |
+
"step": 540
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 3.6333333333333333,
|
| 769 |
+
"grad_norm": 0.05889086052775383,
|
| 770 |
+
"learning_rate": 3.98876404494382e-06,
|
| 771 |
+
"loss": 0.0036,
|
| 772 |
+
"step": 545
|
| 773 |
},
|
| 774 |
{
|
| 775 |
+
"epoch": 3.6666666666666665,
|
| 776 |
+
"grad_norm": 0.03523600473999977,
|
| 777 |
+
"learning_rate": 3.932584269662922e-06,
|
| 778 |
+
"loss": 0.0011,
|
| 779 |
+
"step": 550
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 3.7,
|
| 783 |
+
"grad_norm": 0.03970940411090851,
|
| 784 |
+
"learning_rate": 3.876404494382023e-06,
|
| 785 |
+
"loss": 0.0011,
|
| 786 |
+
"step": 555
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 3.7333333333333334,
|
| 790 |
+
"grad_norm": 0.2314375936985016,
|
| 791 |
+
"learning_rate": 3.820224719101124e-06,
|
| 792 |
+
"loss": 0.0026,
|
| 793 |
+
"step": 560
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"epoch": 3.7666666666666666,
|
| 797 |
+
"grad_norm": 0.07075604051351547,
|
| 798 |
+
"learning_rate": 3.7640449438202247e-06,
|
| 799 |
+
"loss": 0.0031,
|
| 800 |
+
"step": 565
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"epoch": 3.8,
|
| 804 |
+
"grad_norm": 0.09040886908769608,
|
| 805 |
+
"learning_rate": 3.707865168539326e-06,
|
| 806 |
+
"loss": 0.0004,
|
| 807 |
+
"step": 570
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 3.8333333333333335,
|
| 811 |
+
"grad_norm": 0.11175351589918137,
|
| 812 |
+
"learning_rate": 3.651685393258427e-06,
|
| 813 |
+
"loss": 0.0012,
|
| 814 |
+
"step": 575
|
| 815 |
},
|
| 816 |
{
|
| 817 |
+
"epoch": 3.8666666666666667,
|
| 818 |
+
"grad_norm": 0.5862079858779907,
|
| 819 |
+
"learning_rate": 3.595505617977528e-06,
|
| 820 |
+
"loss": 0.001,
|
| 821 |
+
"step": 580
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 3.9,
|
| 825 |
+
"grad_norm": 0.4760936200618744,
|
| 826 |
+
"learning_rate": 3.539325842696629e-06,
|
| 827 |
+
"loss": 0.0008,
|
| 828 |
+
"step": 585
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 3.9333333333333336,
|
| 832 |
+
"grad_norm": 0.18229596316814423,
|
| 833 |
+
"learning_rate": 3.4831460674157306e-06,
|
| 834 |
+
"loss": 0.001,
|
| 835 |
+
"step": 590
|
| 836 |
},
|
| 837 |
{
|
| 838 |
+
"epoch": 3.966666666666667,
|
| 839 |
+
"grad_norm": 0.033338598906993866,
|
| 840 |
+
"learning_rate": 3.4269662921348316e-06,
|
| 841 |
+
"loss": 0.002,
|
| 842 |
+
"step": 595
|
| 843 |
},
|
| 844 |
{
|
| 845 |
+
"epoch": 4.0,
|
| 846 |
+
"grad_norm": 0.19226883351802826,
|
| 847 |
+
"learning_rate": 3.3707865168539327e-06,
|
| 848 |
+
"loss": 0.0006,
|
| 849 |
+
"step": 600
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 4.0,
|
| 853 |
+
"eval_loss": 0.3506176173686981,
|
| 854 |
+
"eval_runtime": 287.3723,
|
| 855 |
+
"eval_samples_per_second": 3.48,
|
| 856 |
"eval_steps_per_second": 0.435,
|
| 857 |
+
"eval_wer": 19.701321079839175,
|
| 858 |
+
"step": 600
|
| 859 |
}
|
| 860 |
],
|
| 861 |
+
"logging_steps": 5,
|
| 862 |
+
"max_steps": 900,
|
| 863 |
"num_input_tokens_seen": 0,
|
| 864 |
+
"num_train_epochs": 6,
|
| 865 |
+
"save_steps": 600,
|
| 866 |
+
"stateful_callbacks": {
|
| 867 |
+
"TrainerControl": {
|
| 868 |
+
"args": {
|
| 869 |
+
"should_epoch_stop": false,
|
| 870 |
+
"should_evaluate": false,
|
| 871 |
+
"should_log": false,
|
| 872 |
+
"should_save": true,
|
| 873 |
+
"should_training_stop": false
|
| 874 |
+
},
|
| 875 |
+
"attributes": {}
|
| 876 |
+
}
|
| 877 |
+
},
|
| 878 |
+
"total_flos": 7.7831995392e+17,
|
| 879 |
+
"train_batch_size": 10,
|
| 880 |
"trial_name": null,
|
| 881 |
"trial_params": null
|
| 882 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db804bff1d34b49eeb93672a0155a7e464e05406a85421985e0ebf298394549d
|
| 3 |
+
size 5240
|