Commit
•
4907e7d
1
Parent(s):
ae7dad9
End of training
Browse files- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +1020 -0
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.98,
|
| 3 |
+
"train_loss": 0.2219778089443193,
|
| 4 |
+
"train_runtime": 55065.7639,
|
| 5 |
+
"train_samples_per_second": 9.607,
|
| 6 |
+
"train_steps_per_second": 0.1
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.98,
|
| 3 |
+
"train_loss": 0.2219778089443193,
|
| 4 |
+
"train_runtime": 55065.7639,
|
| 5 |
+
"train_samples_per_second": 9.607,
|
| 6 |
+
"train_steps_per_second": 0.1
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1020 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.29125073552131653,
|
| 3 |
+
"best_model_checkpoint": "deit-cvc-drop-aug/checkpoint-1800",
|
| 4 |
+
"epoch": 14.979591836734693,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 5505,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.27,
|
| 13 |
+
"learning_rate": 9.991860321746833e-05,
|
| 14 |
+
"loss": 0.5453,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.27,
|
| 19 |
+
"eval_accuracy": 0.7695840775708089,
|
| 20 |
+
"eval_f1": 0.7750933997509339,
|
| 21 |
+
"eval_loss": 0.48243534564971924,
|
| 22 |
+
"eval_precision": 0.7776111944027986,
|
| 23 |
+
"eval_recall": 0.7725918570009931,
|
| 24 |
+
"eval_runtime": 15.4841,
|
| 25 |
+
"eval_samples_per_second": 253.098,
|
| 26 |
+
"eval_steps_per_second": 15.823,
|
| 27 |
+
"step": 100
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.54,
|
| 31 |
+
"learning_rate": 9.967467788732156e-05,
|
| 32 |
+
"loss": 0.4324,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.54,
|
| 37 |
+
"eval_accuracy": 0.7685634090329165,
|
| 38 |
+
"eval_f1": 0.7637405574368324,
|
| 39 |
+
"eval_loss": 0.4796455502510071,
|
| 40 |
+
"eval_precision": 0.8032876712328767,
|
| 41 |
+
"eval_recall": 0.7279046673286991,
|
| 42 |
+
"eval_runtime": 15.2008,
|
| 43 |
+
"eval_samples_per_second": 257.815,
|
| 44 |
+
"eval_steps_per_second": 16.118,
|
| 45 |
+
"step": 200
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.82,
|
| 49 |
+
"learning_rate": 9.926901819904178e-05,
|
| 50 |
+
"loss": 0.4042,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.82,
|
| 55 |
+
"eval_accuracy": 0.8321000255167135,
|
| 56 |
+
"eval_f1": 0.8546819787985865,
|
| 57 |
+
"eval_loss": 0.3790486454963684,
|
| 58 |
+
"eval_precision": 0.7696897374701671,
|
| 59 |
+
"eval_recall": 0.9607745779543198,
|
| 60 |
+
"eval_runtime": 15.77,
|
| 61 |
+
"eval_samples_per_second": 248.509,
|
| 62 |
+
"eval_steps_per_second": 15.536,
|
| 63 |
+
"step": 300
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.09,
|
| 67 |
+
"learning_rate": 9.870294492836612e-05,
|
| 68 |
+
"loss": 0.3849,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.09,
|
| 73 |
+
"eval_accuracy": 0.8101556519520285,
|
| 74 |
+
"eval_f1": 0.8161146811665843,
|
| 75 |
+
"eval_loss": 0.4099680185317993,
|
| 76 |
+
"eval_precision": 0.8125,
|
| 77 |
+
"eval_recall": 0.8197616683217478,
|
| 78 |
+
"eval_runtime": 13.9661,
|
| 79 |
+
"eval_samples_per_second": 280.608,
|
| 80 |
+
"eval_steps_per_second": 17.542,
|
| 81 |
+
"step": 400
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.36,
|
| 85 |
+
"learning_rate": 9.797830113701101e-05,
|
| 86 |
+
"loss": 0.3621,
|
| 87 |
+
"step": 500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.36,
|
| 91 |
+
"eval_accuracy": 0.8387343710130135,
|
| 92 |
+
"eval_f1": 0.8510838831291235,
|
| 93 |
+
"eval_loss": 0.36887437105178833,
|
| 94 |
+
"eval_precision": 0.809865470852018,
|
| 95 |
+
"eval_recall": 0.8967229394240318,
|
| 96 |
+
"eval_runtime": 14.7142,
|
| 97 |
+
"eval_samples_per_second": 266.341,
|
| 98 |
+
"eval_steps_per_second": 16.651,
|
| 99 |
+
"step": 500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.63,
|
| 103 |
+
"learning_rate": 9.709744617190038e-05,
|
| 104 |
+
"loss": 0.3457,
|
| 105 |
+
"step": 600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.63,
|
| 109 |
+
"eval_accuracy": 0.8458790507782598,
|
| 110 |
+
"eval_f1": 0.8642086330935251,
|
| 111 |
+
"eval_loss": 0.3312939703464508,
|
| 112 |
+
"eval_precision": 0.7896466721446179,
|
| 113 |
+
"eval_recall": 0.9543197616683218,
|
| 114 |
+
"eval_runtime": 15.9456,
|
| 115 |
+
"eval_samples_per_second": 245.773,
|
| 116 |
+
"eval_steps_per_second": 15.365,
|
| 117 |
+
"step": 600
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.9,
|
| 121 |
+
"learning_rate": 9.606324798343567e-05,
|
| 122 |
+
"loss": 0.3443,
|
| 123 |
+
"step": 700
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.9,
|
| 127 |
+
"eval_accuracy": 0.840520540954325,
|
| 128 |
+
"eval_f1": 0.8599596683844947,
|
| 129 |
+
"eval_loss": 0.3424055874347687,
|
| 130 |
+
"eval_precision": 0.7835851367905268,
|
| 131 |
+
"eval_recall": 0.9528301886792453,
|
| 132 |
+
"eval_runtime": 15.7574,
|
| 133 |
+
"eval_samples_per_second": 248.708,
|
| 134 |
+
"eval_steps_per_second": 15.548,
|
| 135 |
+
"step": 700
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.18,
|
| 139 |
+
"learning_rate": 9.487907378781853e-05,
|
| 140 |
+
"loss": 0.3287,
|
| 141 |
+
"step": 800
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.18,
|
| 145 |
+
"eval_accuracy": 0.8453687165093136,
|
| 146 |
+
"eval_f1": 0.8560570071258907,
|
| 147 |
+
"eval_loss": 0.33078569173812866,
|
| 148 |
+
"eval_precision": 0.8205828779599271,
|
| 149 |
+
"eval_recall": 0.8947368421052632,
|
| 150 |
+
"eval_runtime": 14.597,
|
| 151 |
+
"eval_samples_per_second": 268.479,
|
| 152 |
+
"eval_steps_per_second": 16.784,
|
| 153 |
+
"step": 800
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.45,
|
| 157 |
+
"learning_rate": 9.35487791038282e-05,
|
| 158 |
+
"loss": 0.3224,
|
| 159 |
+
"step": 900
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.45,
|
| 163 |
+
"eval_accuracy": 0.7655014034192396,
|
| 164 |
+
"eval_f1": 0.7437970448843045,
|
| 165 |
+
"eval_loss": 0.454572468996048,
|
| 166 |
+
"eval_precision": 0.8480610298792117,
|
| 167 |
+
"eval_recall": 0.6623634558093346,
|
| 168 |
+
"eval_runtime": 15.6304,
|
| 169 |
+
"eval_samples_per_second": 250.73,
|
| 170 |
+
"eval_steps_per_second": 15.675,
|
| 171 |
+
"step": 900
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.72,
|
| 175 |
+
"learning_rate": 9.207669519974851e-05,
|
| 176 |
+
"loss": 0.3096,
|
| 177 |
+
"step": 1000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.72,
|
| 181 |
+
"eval_accuracy": 0.8438377137024751,
|
| 182 |
+
"eval_f1": 0.8521024649589173,
|
| 183 |
+
"eval_loss": 0.34022626280784607,
|
| 184 |
+
"eval_precision": 0.8300376647834274,
|
| 185 |
+
"eval_recall": 0.8753723932472691,
|
| 186 |
+
"eval_runtime": 15.4283,
|
| 187 |
+
"eval_samples_per_second": 254.014,
|
| 188 |
+
"eval_steps_per_second": 15.88,
|
| 189 |
+
"step": 1000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.99,
|
| 193 |
+
"learning_rate": 9.046761499131578e-05,
|
| 194 |
+
"loss": 0.3095,
|
| 195 |
+
"step": 1100
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.99,
|
| 199 |
+
"eval_accuracy": 0.8384792038785405,
|
| 200 |
+
"eval_f1": 0.8524131499183958,
|
| 201 |
+
"eval_loss": 0.36910393834114075,
|
| 202 |
+
"eval_precision": 0.8035164835164835,
|
| 203 |
+
"eval_recall": 0.9076464746772592,
|
| 204 |
+
"eval_runtime": 15.4854,
|
| 205 |
+
"eval_samples_per_second": 253.077,
|
| 206 |
+
"eval_steps_per_second": 15.821,
|
| 207 |
+
"step": 1100
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.27,
|
| 211 |
+
"learning_rate": 8.872677743660209e-05,
|
| 212 |
+
"loss": 0.2901,
|
| 213 |
+
"step": 1200
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 3.27,
|
| 217 |
+
"eval_accuracy": 0.8328655269201327,
|
| 218 |
+
"eval_f1": 0.8467119120056167,
|
| 219 |
+
"eval_loss": 0.36432939767837524,
|
| 220 |
+
"eval_precision": 0.8007968127490039,
|
| 221 |
+
"eval_recall": 0.8982125124131083,
|
| 222 |
+
"eval_runtime": 14.6332,
|
| 223 |
+
"eval_samples_per_second": 267.816,
|
| 224 |
+
"eval_steps_per_second": 16.743,
|
| 225 |
+
"step": 1200
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.54,
|
| 229 |
+
"learning_rate": 8.685985047864204e-05,
|
| 230 |
+
"loss": 0.2939,
|
| 231 |
+
"step": 1300
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 3.54,
|
| 235 |
+
"eval_accuracy": 0.8601684103087522,
|
| 236 |
+
"eval_f1": 0.8760180995475113,
|
| 237 |
+
"eval_loss": 0.30208107829093933,
|
| 238 |
+
"eval_precision": 0.8046550290939318,
|
| 239 |
+
"eval_recall": 0.9612711022840119,
|
| 240 |
+
"eval_runtime": 14.6512,
|
| 241 |
+
"eval_samples_per_second": 267.487,
|
| 242 |
+
"eval_steps_per_second": 16.722,
|
| 243 |
+
"step": 1300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.81,
|
| 247 |
+
"learning_rate": 8.487291259133956e-05,
|
| 248 |
+
"loss": 0.2946,
|
| 249 |
+
"step": 1400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 3.81,
|
| 253 |
+
"eval_accuracy": 0.8300586884409288,
|
| 254 |
+
"eval_f1": 0.8342458934793429,
|
| 255 |
+
"eval_loss": 0.36171799898147583,
|
| 256 |
+
"eval_precision": 0.8363273453093812,
|
| 257 |
+
"eval_recall": 0.8321747765640516,
|
| 258 |
+
"eval_runtime": 15.3842,
|
| 259 |
+
"eval_samples_per_second": 254.741,
|
| 260 |
+
"eval_steps_per_second": 15.925,
|
| 261 |
+
"step": 1400
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.08,
|
| 265 |
+
"learning_rate": 8.277243298873936e-05,
|
| 266 |
+
"loss": 0.2856,
|
| 267 |
+
"step": 1500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.08,
|
| 271 |
+
"eval_accuracy": 0.8127073232967594,
|
| 272 |
+
"eval_f1": 0.8116016427104723,
|
| 273 |
+
"eval_loss": 0.48843976855278015,
|
| 274 |
+
"eval_precision": 0.8400637619553666,
|
| 275 |
+
"eval_recall": 0.7850049652432969,
|
| 276 |
+
"eval_runtime": 14.4792,
|
| 277 |
+
"eval_samples_per_second": 270.663,
|
| 278 |
+
"eval_steps_per_second": 16.921,
|
| 279 |
+
"step": 1500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.35,
|
| 283 |
+
"learning_rate": 8.056525056209841e-05,
|
| 284 |
+
"loss": 0.2683,
|
| 285 |
+
"step": 1600
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.35,
|
| 289 |
+
"eval_accuracy": 0.834141362592498,
|
| 290 |
+
"eval_f1": 0.8380667663178873,
|
| 291 |
+
"eval_loss": 0.3540255129337311,
|
| 292 |
+
"eval_precision": 0.841,
|
| 293 |
+
"eval_recall": 0.8351539225422046,
|
| 294 |
+
"eval_runtime": 15.1561,
|
| 295 |
+
"eval_samples_per_second": 258.575,
|
| 296 |
+
"eval_steps_per_second": 16.165,
|
| 297 |
+
"step": 1600
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.63,
|
| 301 |
+
"learning_rate": 7.82585516133363e-05,
|
| 302 |
+
"loss": 0.2724,
|
| 303 |
+
"step": 1700
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.63,
|
| 307 |
+
"eval_accuracy": 0.8581270732329676,
|
| 308 |
+
"eval_f1": 0.8664745437079732,
|
| 309 |
+
"eval_loss": 0.30780917406082153,
|
| 310 |
+
"eval_precision": 0.8390697674418605,
|
| 311 |
+
"eval_recall": 0.8957298907646475,
|
| 312 |
+
"eval_runtime": 14.446,
|
| 313 |
+
"eval_samples_per_second": 271.286,
|
| 314 |
+
"eval_steps_per_second": 16.96,
|
| 315 |
+
"step": 1700
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 4.9,
|
| 319 |
+
"learning_rate": 7.585984645736096e-05,
|
| 320 |
+
"loss": 0.2685,
|
| 321 |
+
"step": 1800
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.9,
|
| 325 |
+
"eval_accuracy": 0.862720081653483,
|
| 326 |
+
"eval_f1": 0.8703614457831326,
|
| 327 |
+
"eval_loss": 0.29125073552131653,
|
| 328 |
+
"eval_precision": 0.8455056179775281,
|
| 329 |
+
"eval_recall": 0.8967229394240318,
|
| 330 |
+
"eval_runtime": 15.5954,
|
| 331 |
+
"eval_samples_per_second": 251.291,
|
| 332 |
+
"eval_steps_per_second": 15.71,
|
| 333 |
+
"step": 1800
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.17,
|
| 337 |
+
"learning_rate": 7.33769449694499e-05,
|
| 338 |
+
"loss": 0.2449,
|
| 339 |
+
"step": 1900
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 5.17,
|
| 343 |
+
"eval_accuracy": 0.8443480479714213,
|
| 344 |
+
"eval_f1": 0.8490099009900989,
|
| 345 |
+
"eval_loss": 0.38658004999160767,
|
| 346 |
+
"eval_precision": 0.8464955577492597,
|
| 347 |
+
"eval_recall": 0.8515392254220456,
|
| 348 |
+
"eval_runtime": 16.0186,
|
| 349 |
+
"eval_samples_per_second": 244.653,
|
| 350 |
+
"eval_steps_per_second": 15.295,
|
| 351 |
+
"step": 1900
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 5.44,
|
| 355 |
+
"learning_rate": 7.081793115730153e-05,
|
| 356 |
+
"loss": 0.2468,
|
| 357 |
+
"step": 2000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.44,
|
| 361 |
+
"eval_accuracy": 0.8588925746363868,
|
| 362 |
+
"eval_f1": 0.8670353450348642,
|
| 363 |
+
"eval_loss": 0.30722111463546753,
|
| 364 |
+
"eval_precision": 0.8405594405594405,
|
| 365 |
+
"eval_recall": 0.8952333664349553,
|
| 366 |
+
"eval_runtime": 15.614,
|
| 367 |
+
"eval_samples_per_second": 250.993,
|
| 368 |
+
"eval_steps_per_second": 15.691,
|
| 369 |
+
"step": 2000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.71,
|
| 373 |
+
"learning_rate": 6.819113684054634e-05,
|
| 374 |
+
"loss": 0.2557,
|
| 375 |
+
"step": 2100
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 5.71,
|
| 379 |
+
"eval_accuracy": 0.8257208471548865,
|
| 380 |
+
"eval_f1": 0.8232858990944373,
|
| 381 |
+
"eval_loss": 0.3735339641571045,
|
| 382 |
+
"eval_precision": 0.8595353862776878,
|
| 383 |
+
"eval_recall": 0.7899702085402185,
|
| 384 |
+
"eval_runtime": 15.378,
|
| 385 |
+
"eval_samples_per_second": 254.845,
|
| 386 |
+
"eval_steps_per_second": 15.932,
|
| 387 |
+
"step": 2100
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 5.99,
|
| 391 |
+
"learning_rate": 6.550511452341459e-05,
|
| 392 |
+
"loss": 0.25,
|
| 393 |
+
"step": 2200
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 5.99,
|
| 397 |
+
"eval_accuracy": 0.8553202347537637,
|
| 398 |
+
"eval_f1": 0.8561278863232682,
|
| 399 |
+
"eval_loss": 0.3116500973701477,
|
| 400 |
+
"eval_precision": 0.8754540736896731,
|
| 401 |
+
"eval_recall": 0.8376365441906654,
|
| 402 |
+
"eval_runtime": 15.151,
|
| 403 |
+
"eval_samples_per_second": 258.663,
|
| 404 |
+
"eval_steps_per_second": 16.171,
|
| 405 |
+
"step": 2200
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 6.26,
|
| 409 |
+
"learning_rate": 6.276860954888322e-05,
|
| 410 |
+
"loss": 0.2256,
|
| 411 |
+
"step": 2300
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 6.26,
|
| 415 |
+
"eval_accuracy": 0.8573615718295483,
|
| 416 |
+
"eval_f1": 0.8652687394552905,
|
| 417 |
+
"eval_loss": 0.3264216184616089,
|
| 418 |
+
"eval_precision": 0.8407494145199064,
|
| 419 |
+
"eval_recall": 0.891261171797418,
|
| 420 |
+
"eval_runtime": 13.8247,
|
| 421 |
+
"eval_samples_per_second": 283.478,
|
| 422 |
+
"eval_steps_per_second": 17.722,
|
| 423 |
+
"step": 2300
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.53,
|
| 427 |
+
"learning_rate": 5.999053162496453e-05,
|
| 428 |
+
"loss": 0.234,
|
| 429 |
+
"step": 2400
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.53,
|
| 433 |
+
"eval_accuracy": 0.8295483541719827,
|
| 434 |
+
"eval_f1": 0.8203335126412049,
|
| 435 |
+
"eval_loss": 0.3617473244667053,
|
| 436 |
+
"eval_precision": 0.8949530516431925,
|
| 437 |
+
"eval_recall": 0.7571996027805362,
|
| 438 |
+
"eval_runtime": 14.3944,
|
| 439 |
+
"eval_samples_per_second": 272.259,
|
| 440 |
+
"eval_steps_per_second": 17.02,
|
| 441 |
+
"step": 2400
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.8,
|
| 445 |
+
"learning_rate": 5.717992581584373e-05,
|
| 446 |
+
"loss": 0.2259,
|
| 447 |
+
"step": 2500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 6.8,
|
| 451 |
+
"eval_accuracy": 0.8609339117121715,
|
| 452 |
+
"eval_f1": 0.8679428156045553,
|
| 453 |
+
"eval_loss": 0.32836005091667175,
|
| 454 |
+
"eval_precision": 0.8476100331282537,
|
| 455 |
+
"eval_recall": 0.8892750744786495,
|
| 456 |
+
"eval_runtime": 15.9578,
|
| 457 |
+
"eval_samples_per_second": 245.585,
|
| 458 |
+
"eval_steps_per_second": 15.353,
|
| 459 |
+
"step": 2500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 7.07,
|
| 463 |
+
"learning_rate": 5.434594309231389e-05,
|
| 464 |
+
"loss": 0.2261,
|
| 465 |
+
"step": 2600
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 7.07,
|
| 469 |
+
"eval_accuracy": 0.8443480479714213,
|
| 470 |
+
"eval_f1": 0.8375066595631327,
|
| 471 |
+
"eval_loss": 0.34862977266311646,
|
| 472 |
+
"eval_precision": 0.903448275862069,
|
| 473 |
+
"eval_recall": 0.7805362462760675,
|
| 474 |
+
"eval_runtime": 14.7094,
|
| 475 |
+
"eval_samples_per_second": 266.428,
|
| 476 |
+
"eval_steps_per_second": 16.656,
|
| 477 |
+
"step": 2600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.35,
|
| 481 |
+
"learning_rate": 5.1497810537392844e-05,
|
| 482 |
+
"loss": 0.2087,
|
| 483 |
+
"step": 2700
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 7.35,
|
| 487 |
+
"eval_accuracy": 0.836948201071702,
|
| 488 |
+
"eval_f1": 0.8365310821181888,
|
| 489 |
+
"eval_loss": 0.3970935046672821,
|
| 490 |
+
"eval_precision": 0.862796833773087,
|
| 491 |
+
"eval_recall": 0.8118172790466733,
|
| 492 |
+
"eval_runtime": 15.0606,
|
| 493 |
+
"eval_samples_per_second": 260.215,
|
| 494 |
+
"eval_steps_per_second": 16.268,
|
| 495 |
+
"step": 2700
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 7.62,
|
| 499 |
+
"learning_rate": 4.8644801304128374e-05,
|
| 500 |
+
"loss": 0.2035,
|
| 501 |
+
"step": 2800
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 7.62,
|
| 505 |
+
"eval_accuracy": 0.878285276856341,
|
| 506 |
+
"eval_f1": 0.8830595734248591,
|
| 507 |
+
"eval_loss": 0.31056222319602966,
|
| 508 |
+
"eval_precision": 0.8721549636803874,
|
| 509 |
+
"eval_recall": 0.894240317775571,
|
| 510 |
+
"eval_runtime": 15.0478,
|
| 511 |
+
"eval_samples_per_second": 260.437,
|
| 512 |
+
"eval_steps_per_second": 16.281,
|
| 513 |
+
"step": 2800
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.89,
|
| 517 |
+
"learning_rate": 4.57962044234053e-05,
|
| 518 |
+
"loss": 0.2116,
|
| 519 |
+
"step": 2900
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 7.89,
|
| 523 |
+
"eval_accuracy": 0.8451135493748405,
|
| 524 |
+
"eval_f1": 0.8428682371214082,
|
| 525 |
+
"eval_loss": 0.3734387159347534,
|
| 526 |
+
"eval_precision": 0.8804759329367225,
|
| 527 |
+
"eval_recall": 0.8083416087388282,
|
| 528 |
+
"eval_runtime": 15.0966,
|
| 529 |
+
"eval_samples_per_second": 259.594,
|
| 530 |
+
"eval_steps_per_second": 16.229,
|
| 531 |
+
"step": 2900
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 8.16,
|
| 535 |
+
"learning_rate": 4.2961294560056445e-05,
|
| 536 |
+
"loss": 0.1956,
|
| 537 |
+
"step": 3000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 8.16,
|
| 541 |
+
"eval_accuracy": 0.85914774177086,
|
| 542 |
+
"eval_f1": 0.863298662704309,
|
| 543 |
+
"eval_loss": 0.3442569077014923,
|
| 544 |
+
"eval_precision": 0.8611660079051383,
|
| 545 |
+
"eval_recall": 0.865441906653426,
|
| 546 |
+
"eval_runtime": 15.4752,
|
| 547 |
+
"eval_samples_per_second": 253.245,
|
| 548 |
+
"eval_steps_per_second": 15.832,
|
| 549 |
+
"step": 3000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 8.44,
|
| 553 |
+
"learning_rate": 4.01493018157476e-05,
|
| 554 |
+
"loss": 0.1826,
|
| 555 |
+
"step": 3100
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 8.44,
|
| 559 |
+
"eval_accuracy": 0.8423067108956367,
|
| 560 |
+
"eval_f1": 0.8373684210526314,
|
| 561 |
+
"eval_loss": 0.3795164227485657,
|
| 562 |
+
"eval_precision": 0.8908174692049272,
|
| 563 |
+
"eval_recall": 0.7899702085402185,
|
| 564 |
+
"eval_runtime": 15.3206,
|
| 565 |
+
"eval_samples_per_second": 255.8,
|
| 566 |
+
"eval_steps_per_second": 15.992,
|
| 567 |
+
"step": 3100
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 8.71,
|
| 571 |
+
"learning_rate": 3.7369381676954284e-05,
|
| 572 |
+
"loss": 0.1918,
|
| 573 |
+
"step": 3200
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 8.71,
|
| 577 |
+
"eval_accuracy": 0.8581270732329676,
|
| 578 |
+
"eval_f1": 0.8569222851260936,
|
| 579 |
+
"eval_loss": 0.33620110154151917,
|
| 580 |
+
"eval_precision": 0.8894230769230769,
|
| 581 |
+
"eval_recall": 0.8267130089374379,
|
| 582 |
+
"eval_runtime": 14.574,
|
| 583 |
+
"eval_samples_per_second": 268.904,
|
| 584 |
+
"eval_steps_per_second": 16.811,
|
| 585 |
+
"step": 3200
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.98,
|
| 589 |
+
"learning_rate": 3.463058520587625e-05,
|
| 590 |
+
"loss": 0.1886,
|
| 591 |
+
"step": 3300
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 8.98,
|
| 595 |
+
"eval_accuracy": 0.8639959173258485,
|
| 596 |
+
"eval_f1": 0.8692666176109884,
|
| 597 |
+
"eval_loss": 0.3259018063545227,
|
| 598 |
+
"eval_precision": 0.8589432864760058,
|
| 599 |
+
"eval_recall": 0.8798411122144985,
|
| 600 |
+
"eval_runtime": 14.5325,
|
| 601 |
+
"eval_samples_per_second": 269.671,
|
| 602 |
+
"eval_steps_per_second": 16.859,
|
| 603 |
+
"step": 3300
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 9.25,
|
| 607 |
+
"learning_rate": 3.194182957134365e-05,
|
| 608 |
+
"loss": 0.1716,
|
| 609 |
+
"step": 3400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 9.25,
|
| 613 |
+
"eval_accuracy": 0.846389385047206,
|
| 614 |
+
"eval_f1": 0.8481331987891019,
|
| 615 |
+
"eval_loss": 0.4269343912601471,
|
| 616 |
+
"eval_precision": 0.862051282051282,
|
| 617 |
+
"eval_recall": 0.8346573982125124,
|
| 618 |
+
"eval_runtime": 15.0706,
|
| 619 |
+
"eval_samples_per_second": 260.043,
|
| 620 |
+
"eval_steps_per_second": 16.257,
|
| 621 |
+
"step": 3400
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 9.52,
|
| 625 |
+
"learning_rate": 2.9311869015663125e-05,
|
| 626 |
+
"loss": 0.1654,
|
| 627 |
+
"step": 3500
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 9.52,
|
| 631 |
+
"eval_accuracy": 0.859658076039806,
|
| 632 |
+
"eval_f1": 0.858974358974359,
|
| 633 |
+
"eval_loss": 0.40663468837738037,
|
| 634 |
+
"eval_precision": 0.8881230116648993,
|
| 635 |
+
"eval_recall": 0.8316782522343595,
|
| 636 |
+
"eval_runtime": 13.6261,
|
| 637 |
+
"eval_samples_per_second": 287.609,
|
| 638 |
+
"eval_steps_per_second": 17.98,
|
| 639 |
+
"step": 3500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 9.8,
|
| 643 |
+
"learning_rate": 2.674926635193134e-05,
|
| 644 |
+
"loss": 0.1625,
|
| 645 |
+
"step": 3600
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 9.8,
|
| 649 |
+
"eval_accuracy": 0.8512375606021945,
|
| 650 |
+
"eval_f1": 0.8488462535649469,
|
| 651 |
+
"eval_loss": 0.3926689624786377,
|
| 652 |
+
"eval_precision": 0.8882257189365166,
|
| 653 |
+
"eval_recall": 0.8128103277060575,
|
| 654 |
+
"eval_runtime": 15.0764,
|
| 655 |
+
"eval_samples_per_second": 259.943,
|
| 656 |
+
"eval_steps_per_second": 16.251,
|
| 657 |
+
"step": 3600
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 10.07,
|
| 661 |
+
"learning_rate": 2.42623650846177e-05,
|
| 662 |
+
"loss": 0.1659,
|
| 663 |
+
"step": 3700
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 10.07,
|
| 667 |
+
"eval_accuracy": 0.8548099004848175,
|
| 668 |
+
"eval_f1": 0.8529335745670716,
|
| 669 |
+
"eval_loss": 0.37967267632484436,
|
| 670 |
+
"eval_precision": 0.889487870619946,
|
| 671 |
+
"eval_recall": 0.8192651439920556,
|
| 672 |
+
"eval_runtime": 14.8402,
|
| 673 |
+
"eval_samples_per_second": 264.08,
|
| 674 |
+
"eval_steps_per_second": 16.509,
|
| 675 |
+
"step": 3700
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 10.34,
|
| 679 |
+
"learning_rate": 2.1859262244187556e-05,
|
| 680 |
+
"loss": 0.1519,
|
| 681 |
+
"step": 3800
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 10.34,
|
| 685 |
+
"eval_accuracy": 0.8512375606021945,
|
| 686 |
+
"eval_f1": 0.8502440277421012,
|
| 687 |
+
"eval_loss": 0.4088890254497528,
|
| 688 |
+
"eval_precision": 0.8807876530069185,
|
| 689 |
+
"eval_recall": 0.8217477656405164,
|
| 690 |
+
"eval_runtime": 15.4244,
|
| 691 |
+
"eval_samples_per_second": 254.079,
|
| 692 |
+
"eval_steps_per_second": 15.884,
|
| 693 |
+
"step": 3800
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 10.61,
|
| 697 |
+
"learning_rate": 1.9547782024213047e-05,
|
| 698 |
+
"loss": 0.1484,
|
| 699 |
+
"step": 3900
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 10.61,
|
| 703 |
+
"eval_accuracy": 0.8545547333503445,
|
| 704 |
+
"eval_f1": 0.8533950617283951,
|
| 705 |
+
"eval_loss": 0.3864934742450714,
|
| 706 |
+
"eval_precision": 0.8852721451440768,
|
| 707 |
+
"eval_recall": 0.823733862959285,
|
| 708 |
+
"eval_runtime": 14.9541,
|
| 709 |
+
"eval_samples_per_second": 262.069,
|
| 710 |
+
"eval_steps_per_second": 16.383,
|
| 711 |
+
"step": 3900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 10.88,
|
| 715 |
+
"learning_rate": 1.7335450306805827e-05,
|
| 716 |
+
"loss": 0.1427,
|
| 717 |
+
"step": 4000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 10.88,
|
| 721 |
+
"eval_accuracy": 0.846389385047206,
|
| 722 |
+
"eval_f1": 0.8427377220480669,
|
| 723 |
+
"eval_loss": 0.4346730411052704,
|
| 724 |
+
"eval_precision": 0.8891951488423374,
|
| 725 |
+
"eval_recall": 0.8008937437934459,
|
| 726 |
+
"eval_runtime": 15.9395,
|
| 727 |
+
"eval_samples_per_second": 245.867,
|
| 728 |
+
"eval_steps_per_second": 15.371,
|
| 729 |
+
"step": 4000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 11.16,
|
| 733 |
+
"learning_rate": 1.522947015931348e-05,
|
| 734 |
+
"loss": 0.1375,
|
| 735 |
+
"step": 4100
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 11.16,
|
| 739 |
+
"eval_accuracy": 0.8548099004848175,
|
| 740 |
+
"eval_f1": 0.8532370389476398,
|
| 741 |
+
"eval_loss": 0.4687642753124237,
|
| 742 |
+
"eval_precision": 0.8878153515834675,
|
| 743 |
+
"eval_recall": 0.8212512413108243,
|
| 744 |
+
"eval_runtime": 14.957,
|
| 745 |
+
"eval_samples_per_second": 262.017,
|
| 746 |
+
"eval_steps_per_second": 16.38,
|
| 747 |
+
"step": 4100
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 11.43,
|
| 751 |
+
"learning_rate": 1.3236698382059287e-05,
|
| 752 |
+
"loss": 0.1276,
|
| 753 |
+
"step": 4200
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 11.43,
|
| 757 |
+
"eval_accuracy": 0.846899719316152,
|
| 758 |
+
"eval_f1": 0.8426023084994753,
|
| 759 |
+
"eval_loss": 0.4686568081378937,
|
| 760 |
+
"eval_precision": 0.8932146829810901,
|
| 761 |
+
"eval_recall": 0.7974180734856008,
|
| 762 |
+
"eval_runtime": 13.9054,
|
| 763 |
+
"eval_samples_per_second": 281.833,
|
| 764 |
+
"eval_steps_per_second": 17.619,
|
| 765 |
+
"step": 4200
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 11.7,
|
| 769 |
+
"learning_rate": 1.1363623183482775e-05,
|
| 770 |
+
"loss": 0.1275,
|
| 771 |
+
"step": 4300
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 11.7,
|
| 775 |
+
"eval_accuracy": 0.8486858892574637,
|
| 776 |
+
"eval_f1": 0.8447237496726892,
|
| 777 |
+
"eval_loss": 0.4493071436882019,
|
| 778 |
+
"eval_precision": 0.8936288088642659,
|
| 779 |
+
"eval_recall": 0.8008937437934459,
|
| 780 |
+
"eval_runtime": 14.5625,
|
| 781 |
+
"eval_samples_per_second": 269.116,
|
| 782 |
+
"eval_steps_per_second": 16.824,
|
| 783 |
+
"step": 4300
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 11.97,
|
| 787 |
+
"learning_rate": 9.616343055368083e-06,
|
| 788 |
+
"loss": 0.1349,
|
| 789 |
+
"step": 4400
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 11.97,
|
| 793 |
+
"eval_accuracy": 0.8423067108956367,
|
| 794 |
+
"eval_f1": 0.8360742705570292,
|
| 795 |
+
"eval_loss": 0.4618338346481323,
|
| 796 |
+
"eval_precision": 0.89749430523918,
|
| 797 |
+
"eval_recall": 0.7825223435948362,
|
| 798 |
+
"eval_runtime": 14.9603,
|
| 799 |
+
"eval_samples_per_second": 261.96,
|
| 800 |
+
"eval_steps_per_second": 16.377,
|
| 801 |
+
"step": 4400
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 12.24,
|
| 805 |
+
"learning_rate": 8.000546916939644e-06,
|
| 806 |
+
"loss": 0.1217,
|
| 807 |
+
"step": 4500
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 12.24,
|
| 811 |
+
"eval_accuracy": 0.849706557795356,
|
| 812 |
+
"eval_f1": 0.8450407787424362,
|
| 813 |
+
"eval_loss": 0.4635533094406128,
|
| 814 |
+
"eval_precision": 0.8987129266927812,
|
| 815 |
+
"eval_recall": 0.7974180734856008,
|
| 816 |
+
"eval_runtime": 15.9792,
|
| 817 |
+
"eval_samples_per_second": 245.256,
|
| 818 |
+
"eval_steps_per_second": 15.332,
|
| 819 |
+
"step": 4500
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 12.52,
|
| 823 |
+
"learning_rate": 6.521495592473259e-06,
|
| 824 |
+
"loss": 0.1211,
|
| 825 |
+
"step": 4600
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 12.52,
|
| 829 |
+
"eval_accuracy": 0.8555754018882368,
|
| 830 |
+
"eval_f1": 0.855316973415133,
|
| 831 |
+
"eval_loss": 0.45266029238700867,
|
| 832 |
+
"eval_precision": 0.8814541622760801,
|
| 833 |
+
"eval_recall": 0.8306852035749752,
|
| 834 |
+
"eval_runtime": 15.4568,
|
| 835 |
+
"eval_samples_per_second": 253.545,
|
| 836 |
+
"eval_steps_per_second": 15.851,
|
| 837 |
+
"step": 4600
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 12.79,
|
| 841 |
+
"learning_rate": 5.184004682729348e-06,
|
| 842 |
+
"loss": 0.1164,
|
| 843 |
+
"step": 4700
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 12.79,
|
| 847 |
+
"eval_accuracy": 0.8545547333503445,
|
| 848 |
+
"eval_f1": 0.851639770952629,
|
| 849 |
+
"eval_loss": 0.46692270040512085,
|
| 850 |
+
"eval_precision": 0.8949671772428884,
|
| 851 |
+
"eval_recall": 0.8123138033763655,
|
| 852 |
+
"eval_runtime": 16.339,
|
| 853 |
+
"eval_samples_per_second": 239.856,
|
| 854 |
+
"eval_steps_per_second": 14.995,
|
| 855 |
+
"step": 4700
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 13.06,
|
| 859 |
+
"learning_rate": 3.992428885976652e-06,
|
| 860 |
+
"loss": 0.1119,
|
| 861 |
+
"step": 4800
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 13.06,
|
| 865 |
+
"eval_accuracy": 0.8517478948711406,
|
| 866 |
+
"eval_f1": 0.8495987574424021,
|
| 867 |
+
"eval_loss": 0.461725115776062,
|
| 868 |
+
"eval_precision": 0.8875067604110329,
|
| 869 |
+
"eval_recall": 0.8147964250248262,
|
| 870 |
+
"eval_runtime": 15.7126,
|
| 871 |
+
"eval_samples_per_second": 249.418,
|
| 872 |
+
"eval_steps_per_second": 15.593,
|
| 873 |
+
"step": 4800
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 13.33,
|
| 877 |
+
"learning_rate": 2.9506478196551055e-06,
|
| 878 |
+
"loss": 0.11,
|
| 879 |
+
"step": 4900
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 13.33,
|
| 883 |
+
"eval_accuracy": 0.8507272263332483,
|
| 884 |
+
"eval_f1": 0.848012470771629,
|
| 885 |
+
"eval_loss": 0.4718102216720581,
|
| 886 |
+
"eval_precision": 0.8893732970027248,
|
| 887 |
+
"eval_recall": 0.8103277060575969,
|
| 888 |
+
"eval_runtime": 15.1226,
|
| 889 |
+
"eval_samples_per_second": 259.149,
|
| 890 |
+
"eval_steps_per_second": 16.201,
|
| 891 |
+
"step": 4900
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 13.61,
|
| 895 |
+
"learning_rate": 2.062053388840768e-06,
|
| 896 |
+
"loss": 0.1138,
|
| 897 |
+
"step": 5000
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 13.61,
|
| 901 |
+
"eval_accuracy": 0.8479203878540444,
|
| 902 |
+
"eval_f1": 0.8437336130047195,
|
| 903 |
+
"eval_loss": 0.48920777440071106,
|
| 904 |
+
"eval_precision": 0.8938888888888888,
|
| 905 |
+
"eval_recall": 0.7989076464746773,
|
| 906 |
+
"eval_runtime": 15.5429,
|
| 907 |
+
"eval_samples_per_second": 252.14,
|
| 908 |
+
"eval_steps_per_second": 15.763,
|
| 909 |
+
"step": 5000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 13.88,
|
| 913 |
+
"learning_rate": 1.329538742639358e-06,
|
| 914 |
+
"loss": 0.1058,
|
| 915 |
+
"step": 5100
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 13.88,
|
| 919 |
+
"eval_accuracy": 0.849961724929829,
|
| 920 |
+
"eval_f1": 0.8474312402698495,
|
| 921 |
+
"eval_loss": 0.47247427701950073,
|
| 922 |
+
"eval_precision": 0.8875,
|
| 923 |
+
"eval_recall": 0.810824230387289,
|
| 924 |
+
"eval_runtime": 16.0011,
|
| 925 |
+
"eval_samples_per_second": 244.92,
|
| 926 |
+
"eval_steps_per_second": 15.311,
|
| 927 |
+
"step": 5100
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 14.15,
|
| 931 |
+
"learning_rate": 7.554888544652305e-07,
|
| 932 |
+
"loss": 0.1042,
|
| 933 |
+
"step": 5200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 14.15,
|
| 937 |
+
"eval_accuracy": 0.849706557795356,
|
| 938 |
+
"eval_f1": 0.8464946572843368,
|
| 939 |
+
"eval_loss": 0.47878143191337585,
|
| 940 |
+
"eval_precision": 0.8908392759188152,
|
| 941 |
+
"eval_recall": 0.8063555114200596,
|
| 942 |
+
"eval_runtime": 15.3441,
|
| 943 |
+
"eval_samples_per_second": 255.408,
|
| 944 |
+
"eval_steps_per_second": 15.967,
|
| 945 |
+
"step": 5200
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 14.42,
|
| 949 |
+
"learning_rate": 3.4177275687476974e-07,
|
| 950 |
+
"loss": 0.107,
|
| 951 |
+
"step": 5300
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 14.42,
|
| 955 |
+
"eval_accuracy": 0.849706557795356,
|
| 956 |
+
"eval_f1": 0.8466545170528509,
|
| 957 |
+
"eval_loss": 0.4759483337402344,
|
| 958 |
+
"eval_precision": 0.8899835796387521,
|
| 959 |
+
"eval_recall": 0.8073485600794439,
|
| 960 |
+
"eval_runtime": 15.151,
|
| 961 |
+
"eval_samples_per_second": 258.663,
|
| 962 |
+
"eval_steps_per_second": 16.171,
|
| 963 |
+
"step": 5300
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 14.69,
|
| 967 |
+
"learning_rate": 8.973745623699903e-08,
|
| 968 |
+
"loss": 0.1047,
|
| 969 |
+
"step": 5400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 14.69,
|
| 973 |
+
"eval_accuracy": 0.8489410563919367,
|
| 974 |
+
"eval_f1": 0.8458333333333334,
|
| 975 |
+
"eval_loss": 0.4766782820224762,
|
| 976 |
+
"eval_precision": 0.8893756845564075,
|
| 977 |
+
"eval_recall": 0.8063555114200596,
|
| 978 |
+
"eval_runtime": 15.3016,
|
| 979 |
+
"eval_samples_per_second": 256.118,
|
| 980 |
+
"eval_steps_per_second": 16.011,
|
| 981 |
+
"step": 5400
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 14.97,
|
| 985 |
+
"learning_rate": 2.0354705417280351e-10,
|
| 986 |
+
"loss": 0.1085,
|
| 987 |
+
"step": 5500
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 14.97,
|
| 991 |
+
"eval_accuracy": 0.8489410563919367,
|
| 992 |
+
"eval_f1": 0.8458333333333334,
|
| 993 |
+
"eval_loss": 0.47687891125679016,
|
| 994 |
+
"eval_precision": 0.8893756845564075,
|
| 995 |
+
"eval_recall": 0.8063555114200596,
|
| 996 |
+
"eval_runtime": 15.647,
|
| 997 |
+
"eval_samples_per_second": 250.463,
|
| 998 |
+
"eval_steps_per_second": 15.658,
|
| 999 |
+
"step": 5500
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 14.98,
|
| 1003 |
+
"step": 5505,
|
| 1004 |
+
"total_flos": 2.046979800964418e+20,
|
| 1005 |
+
"train_loss": 0.2219778089443193,
|
| 1006 |
+
"train_runtime": 55065.7639,
|
| 1007 |
+
"train_samples_per_second": 9.607,
|
| 1008 |
+
"train_steps_per_second": 0.1
|
| 1009 |
+
}
|
| 1010 |
+
],
|
| 1011 |
+
"logging_steps": 100,
|
| 1012 |
+
"max_steps": 5505,
|
| 1013 |
+
"num_input_tokens_seen": 0,
|
| 1014 |
+
"num_train_epochs": 15,
|
| 1015 |
+
"save_steps": 100,
|
| 1016 |
+
"total_flos": 2.046979800964418e+20,
|
| 1017 |
+
"train_batch_size": 16,
|
| 1018 |
+
"trial_name": null,
|
| 1019 |
+
"trial_params": null
|
| 1020 |
+
}
|