Commit
•
cba4cc3
1
Parent(s):
4004a5d
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +517 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"eval_loss": 0.5975067615509033,
|
| 4 |
+
"eval_runtime": 13.5079,
|
| 5 |
+
"eval_samples": 375,
|
| 6 |
+
"eval_samples_per_second": 27.761,
|
| 7 |
+
"eval_steps_per_second": 0.888,
|
| 8 |
+
"eval_wer": 0.4980947196516059,
|
| 9 |
+
"train_loss": 1.1891312944889068,
|
| 10 |
+
"train_runtime": 10628.5861,
|
| 11 |
+
"train_samples": 1008,
|
| 12 |
+
"train_samples_per_second": 18.968,
|
| 13 |
+
"train_steps_per_second": 0.602
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"eval_loss": 0.5975067615509033,
|
| 4 |
+
"eval_runtime": 13.5079,
|
| 5 |
+
"eval_samples": 375,
|
| 6 |
+
"eval_samples_per_second": 27.761,
|
| 7 |
+
"eval_steps_per_second": 0.888,
|
| 8 |
+
"eval_wer": 0.4980947196516059
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262058993
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3693fc1847b8bdaebf9ee28b06bc7e686ab21ab7953c3892579c91f5b0caf21d
|
| 3 |
size 1262058993
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"train_loss": 1.1891312944889068,
|
| 4 |
+
"train_runtime": 10628.5861,
|
| 5 |
+
"train_samples": 1008,
|
| 6 |
+
"train_samples_per_second": 18.968,
|
| 7 |
+
"train_steps_per_second": 0.602
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,517 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 200.0,
|
| 5 |
+
"global_step": 6400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 3.12,
|
| 12 |
+
"learning_rate": 1.3439999999999998e-05,
|
| 13 |
+
"loss": 22.3442,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 6.25,
|
| 18 |
+
"learning_rate": 2.7439999999999998e-05,
|
| 19 |
+
"loss": 5.9373,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 9.38,
|
| 24 |
+
"learning_rate": 4.1439999999999996e-05,
|
| 25 |
+
"loss": 3.7179,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 12.5,
|
| 30 |
+
"learning_rate": 5.544e-05,
|
| 31 |
+
"loss": 3.0789,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 15.62,
|
| 36 |
+
"learning_rate": 6.944e-05,
|
| 37 |
+
"loss": 2.9287,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 15.62,
|
| 42 |
+
"eval_loss": 2.877366304397583,
|
| 43 |
+
"eval_runtime": 13.8367,
|
| 44 |
+
"eval_samples_per_second": 27.102,
|
| 45 |
+
"eval_steps_per_second": 0.867,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 18.75,
|
| 51 |
+
"learning_rate": 6.886101694915253e-05,
|
| 52 |
+
"loss": 2.8183,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 21.88,
|
| 57 |
+
"learning_rate": 6.767457627118643e-05,
|
| 58 |
+
"loss": 2.4012,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 25.0,
|
| 63 |
+
"learning_rate": 6.648813559322033e-05,
|
| 64 |
+
"loss": 1.6359,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 28.12,
|
| 69 |
+
"learning_rate": 6.530169491525422e-05,
|
| 70 |
+
"loss": 1.2603,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 31.25,
|
| 75 |
+
"learning_rate": 6.411525423728813e-05,
|
| 76 |
+
"loss": 1.1182,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 31.25,
|
| 81 |
+
"eval_loss": 0.6247776746749878,
|
| 82 |
+
"eval_runtime": 13.8909,
|
| 83 |
+
"eval_samples_per_second": 26.996,
|
| 84 |
+
"eval_steps_per_second": 0.864,
|
| 85 |
+
"eval_wer": 0.713119216113228,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 34.38,
|
| 90 |
+
"learning_rate": 6.292881355932203e-05,
|
| 91 |
+
"loss": 1.0279,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 37.5,
|
| 96 |
+
"learning_rate": 6.174237288135593e-05,
|
| 97 |
+
"loss": 0.9694,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 40.62,
|
| 102 |
+
"learning_rate": 6.0555932203389824e-05,
|
| 103 |
+
"loss": 0.8993,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 43.75,
|
| 108 |
+
"learning_rate": 5.936949152542372e-05,
|
| 109 |
+
"loss": 0.8663,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 46.88,
|
| 114 |
+
"learning_rate": 5.818305084745762e-05,
|
| 115 |
+
"loss": 0.8329,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 46.88,
|
| 120 |
+
"eval_loss": 0.5573042631149292,
|
| 121 |
+
"eval_runtime": 13.9992,
|
| 122 |
+
"eval_samples_per_second": 26.787,
|
| 123 |
+
"eval_steps_per_second": 0.857,
|
| 124 |
+
"eval_wer": 0.5792052259118128,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 50.0,
|
| 129 |
+
"learning_rate": 5.6996610169491526e-05,
|
| 130 |
+
"loss": 0.8219,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 53.12,
|
| 135 |
+
"learning_rate": 5.581016949152542e-05,
|
| 136 |
+
"loss": 0.8078,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 56.25,
|
| 141 |
+
"learning_rate": 5.4623728813559317e-05,
|
| 142 |
+
"loss": 0.7718,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 59.38,
|
| 147 |
+
"learning_rate": 5.3437288135593215e-05,
|
| 148 |
+
"loss": 0.7365,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 62.5,
|
| 153 |
+
"learning_rate": 5.2250847457627114e-05,
|
| 154 |
+
"loss": 0.7109,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 62.5,
|
| 159 |
+
"eval_loss": 0.54203200340271,
|
| 160 |
+
"eval_runtime": 13.6769,
|
| 161 |
+
"eval_samples_per_second": 27.419,
|
| 162 |
+
"eval_steps_per_second": 0.877,
|
| 163 |
+
"eval_wer": 0.568317909635275,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 65.62,
|
| 168 |
+
"learning_rate": 5.106440677966102e-05,
|
| 169 |
+
"loss": 0.691,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 68.75,
|
| 174 |
+
"learning_rate": 4.987796610169491e-05,
|
| 175 |
+
"loss": 0.69,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 71.88,
|
| 180 |
+
"learning_rate": 4.870338983050847e-05,
|
| 181 |
+
"loss": 0.6662,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 75.0,
|
| 186 |
+
"learning_rate": 4.7528813559322025e-05,
|
| 187 |
+
"loss": 0.6387,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 78.12,
|
| 192 |
+
"learning_rate": 4.634237288135593e-05,
|
| 193 |
+
"loss": 0.6295,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 78.12,
|
| 198 |
+
"eval_loss": 0.5166199803352356,
|
| 199 |
+
"eval_runtime": 13.9794,
|
| 200 |
+
"eval_samples_per_second": 26.825,
|
| 201 |
+
"eval_steps_per_second": 0.858,
|
| 202 |
+
"eval_wer": 0.5394665215024497,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 81.25,
|
| 207 |
+
"learning_rate": 4.515593220338983e-05,
|
| 208 |
+
"loss": 0.6157,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 84.38,
|
| 213 |
+
"learning_rate": 4.396949152542372e-05,
|
| 214 |
+
"loss": 0.596,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 87.5,
|
| 219 |
+
"learning_rate": 4.2783050847457626e-05,
|
| 220 |
+
"loss": 0.5883,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 90.62,
|
| 225 |
+
"learning_rate": 4.159661016949152e-05,
|
| 226 |
+
"loss": 0.5843,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 93.75,
|
| 231 |
+
"learning_rate": 4.041016949152542e-05,
|
| 232 |
+
"loss": 0.5715,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 93.75,
|
| 237 |
+
"eval_loss": 0.5486962795257568,
|
| 238 |
+
"eval_runtime": 13.6771,
|
| 239 |
+
"eval_samples_per_second": 27.418,
|
| 240 |
+
"eval_steps_per_second": 0.877,
|
| 241 |
+
"eval_wer": 0.562874251497006,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 96.88,
|
| 246 |
+
"learning_rate": 3.922372881355932e-05,
|
| 247 |
+
"loss": 0.5581,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 100.0,
|
| 252 |
+
"learning_rate": 3.8037288135593213e-05,
|
| 253 |
+
"loss": 0.5262,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 103.12,
|
| 258 |
+
"learning_rate": 3.685084745762712e-05,
|
| 259 |
+
"loss": 0.5337,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 106.25,
|
| 264 |
+
"learning_rate": 3.566440677966101e-05,
|
| 265 |
+
"loss": 0.5229,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 109.38,
|
| 270 |
+
"learning_rate": 3.4477966101694916e-05,
|
| 271 |
+
"loss": 0.5016,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 109.38,
|
| 276 |
+
"eval_loss": 0.5369759202003479,
|
| 277 |
+
"eval_runtime": 14.0144,
|
| 278 |
+
"eval_samples_per_second": 26.758,
|
| 279 |
+
"eval_steps_per_second": 0.856,
|
| 280 |
+
"eval_wer": 0.5470876428960262,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 112.5,
|
| 285 |
+
"learning_rate": 3.329152542372881e-05,
|
| 286 |
+
"loss": 0.4817,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 115.62,
|
| 291 |
+
"learning_rate": 3.2105084745762706e-05,
|
| 292 |
+
"loss": 0.4959,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 118.75,
|
| 297 |
+
"learning_rate": 3.0918644067796605e-05,
|
| 298 |
+
"loss": 0.4786,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 121.88,
|
| 303 |
+
"learning_rate": 2.9732203389830507e-05,
|
| 304 |
+
"loss": 0.4671,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 125.0,
|
| 309 |
+
"learning_rate": 2.8545762711864405e-05,
|
| 310 |
+
"loss": 0.4661,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 125.0,
|
| 315 |
+
"eval_loss": 0.5621253252029419,
|
| 316 |
+
"eval_runtime": 13.7208,
|
| 317 |
+
"eval_samples_per_second": 27.331,
|
| 318 |
+
"eval_steps_per_second": 0.875,
|
| 319 |
+
"eval_wer": 0.5394665215024497,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 128.12,
|
| 324 |
+
"learning_rate": 2.7359322033898304e-05,
|
| 325 |
+
"loss": 0.4538,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 131.25,
|
| 330 |
+
"learning_rate": 2.61728813559322e-05,
|
| 331 |
+
"loss": 0.4482,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 134.38,
|
| 336 |
+
"learning_rate": 2.4986440677966097e-05,
|
| 337 |
+
"loss": 0.4518,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 137.5,
|
| 342 |
+
"learning_rate": 2.38e-05,
|
| 343 |
+
"loss": 0.4371,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 140.62,
|
| 348 |
+
"learning_rate": 2.2613559322033898e-05,
|
| 349 |
+
"loss": 0.423,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 140.62,
|
| 354 |
+
"eval_loss": 0.5657923221588135,
|
| 355 |
+
"eval_runtime": 13.7521,
|
| 356 |
+
"eval_samples_per_second": 27.269,
|
| 357 |
+
"eval_steps_per_second": 0.873,
|
| 358 |
+
"eval_wer": 0.5247686445291235,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 143.75,
|
| 363 |
+
"learning_rate": 2.1427118644067796e-05,
|
| 364 |
+
"loss": 0.4134,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 146.88,
|
| 369 |
+
"learning_rate": 2.024067796610169e-05,
|
| 370 |
+
"loss": 0.4014,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 150.0,
|
| 375 |
+
"learning_rate": 1.905423728813559e-05,
|
| 376 |
+
"loss": 0.4001,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 153.12,
|
| 381 |
+
"learning_rate": 1.786779661016949e-05,
|
| 382 |
+
"loss": 0.3777,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 156.25,
|
| 387 |
+
"learning_rate": 1.6681355932203387e-05,
|
| 388 |
+
"loss": 0.3793,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 156.25,
|
| 393 |
+
"eval_loss": 0.5920763611793518,
|
| 394 |
+
"eval_runtime": 13.612,
|
| 395 |
+
"eval_samples_per_second": 27.549,
|
| 396 |
+
"eval_steps_per_second": 0.882,
|
| 397 |
+
"eval_wer": 0.4980947196516059,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 159.38,
|
| 402 |
+
"learning_rate": 1.549491525423729e-05,
|
| 403 |
+
"loss": 0.3843,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 162.5,
|
| 408 |
+
"learning_rate": 1.4308474576271186e-05,
|
| 409 |
+
"loss": 0.3918,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 165.62,
|
| 414 |
+
"learning_rate": 1.3122033898305082e-05,
|
| 415 |
+
"loss": 0.3593,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 168.75,
|
| 420 |
+
"learning_rate": 1.1935593220338983e-05,
|
| 421 |
+
"loss": 0.3674,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 171.88,
|
| 426 |
+
"learning_rate": 1.074915254237288e-05,
|
| 427 |
+
"loss": 0.3651,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 171.88,
|
| 432 |
+
"eval_loss": 0.5986579060554504,
|
| 433 |
+
"eval_runtime": 13.8361,
|
| 434 |
+
"eval_samples_per_second": 27.103,
|
| 435 |
+
"eval_steps_per_second": 0.867,
|
| 436 |
+
"eval_wer": 0.48884050081654873,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 175.0,
|
| 441 |
+
"learning_rate": 9.56271186440678e-06,
|
| 442 |
+
"loss": 0.345,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 178.12,
|
| 447 |
+
"learning_rate": 8.376271186440677e-06,
|
| 448 |
+
"loss": 0.3527,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 181.25,
|
| 453 |
+
"learning_rate": 7.189830508474576e-06,
|
| 454 |
+
"loss": 0.3446,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 184.38,
|
| 459 |
+
"learning_rate": 6.003389830508474e-06,
|
| 460 |
+
"loss": 0.3461,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 187.5,
|
| 465 |
+
"learning_rate": 4.816949152542372e-06,
|
| 466 |
+
"loss": 0.3351,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 187.5,
|
| 471 |
+
"eval_loss": 0.6017099022865295,
|
| 472 |
+
"eval_runtime": 13.6253,
|
| 473 |
+
"eval_samples_per_second": 27.522,
|
| 474 |
+
"eval_steps_per_second": 0.881,
|
| 475 |
+
"eval_wer": 0.49482852476864453,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 190.62,
|
| 480 |
+
"learning_rate": 3.630508474576271e-06,
|
| 481 |
+
"loss": 0.3368,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 193.75,
|
| 486 |
+
"learning_rate": 2.444067796610169e-06,
|
| 487 |
+
"loss": 0.3405,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 196.88,
|
| 492 |
+
"learning_rate": 1.2576271186440677e-06,
|
| 493 |
+
"loss": 0.3271,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 200.0,
|
| 498 |
+
"learning_rate": 7.118644067796609e-08,
|
| 499 |
+
"loss": 0.334,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 200.0,
|
| 504 |
+
"step": 6400,
|
| 505 |
+
"total_flos": 2.9581900723233317e+19,
|
| 506 |
+
"train_loss": 1.1891312944889068,
|
| 507 |
+
"train_runtime": 10628.5861,
|
| 508 |
+
"train_samples_per_second": 18.968,
|
| 509 |
+
"train_steps_per_second": 0.602
|
| 510 |
+
}
|
| 511 |
+
],
|
| 512 |
+
"max_steps": 6400,
|
| 513 |
+
"num_train_epochs": 200,
|
| 514 |
+
"total_flos": 2.9581900723233317e+19,
|
| 515 |
+
"trial_name": null,
|
| 516 |
+
"trial_params": null
|
| 517 |
+
}
|