carted-nlp/categorization-finetuned-20220721-164940-distilled-20220811-013354
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.8775628491620112,
|
| 4 |
+
"eval_f1": 0.8768335428087971,
|
| 5 |
+
"eval_loss": 0.06446786969900131,
|
| 6 |
+
"eval_runtime": 36.0103,
|
| 7 |
+
"eval_samples_per_second": 3976.641,
|
| 8 |
+
"eval_steps_per_second": 41.433,
|
| 9 |
+
"test_samples": 143200,
|
| 10 |
+
"train_loss": 0.06639543622843883,
|
| 11 |
+
"train_runtime": 19911.3581,
|
| 12 |
+
"train_samples": 1138117,
|
| 13 |
+
"train_samples_per_second": 1714.776,
|
| 14 |
+
"train_steps_per_second": 6.699
|
| 15 |
+
}
|
runs/Aug11_01-34-27_product-categorization/events.out.tfevents.1660201727.product-categorization.2923.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70a1274eb11cc975ac731a8bc855ff76898e266004b91e8620ed9413333c410f
|
| 3 |
+
size 416
|
test_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.8775628491620112,
|
| 4 |
+
"eval_f1": 0.8768335428087971,
|
| 5 |
+
"eval_loss": 0.06446786969900131,
|
| 6 |
+
"eval_runtime": 36.0103,
|
| 7 |
+
"eval_samples_per_second": 3976.641,
|
| 8 |
+
"eval_steps_per_second": 41.433,
|
| 9 |
+
"test_samples": 143200
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"train_loss": 0.06639543622843883,
|
| 4 |
+
"train_runtime": 19911.3581,
|
| 5 |
+
"train_samples": 1138117,
|
| 6 |
+
"train_samples_per_second": 1714.776,
|
| 7 |
+
"train_steps_per_second": 6.699
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,873 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.06449923664331436,
|
| 3 |
+
"best_model_checkpoint": "./categorization-finetuned-20220721-164940-distilled-20220811-013354/checkpoint-127500",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"global_step": 133380,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.56,
|
| 12 |
+
"learning_rate": 1.999716279095759e-05,
|
| 13 |
+
"loss": 0.2702,
|
| 14 |
+
"step": 2500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.56,
|
| 18 |
+
"eval_accuracy": 0.7832328973737684,
|
| 19 |
+
"eval_f1": 0.7782592459057264,
|
| 20 |
+
"eval_loss": 0.12900541722774506,
|
| 21 |
+
"eval_runtime": 33.288,
|
| 22 |
+
"eval_samples_per_second": 4302.123,
|
| 23 |
+
"eval_steps_per_second": 44.821,
|
| 24 |
+
"step": 2500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.12,
|
| 28 |
+
"learning_rate": 1.9965262675626726e-05,
|
| 29 |
+
"loss": 0.1246,
|
| 30 |
+
"step": 5000
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.12,
|
| 34 |
+
"eval_accuracy": 0.8168830171288117,
|
| 35 |
+
"eval_f1": 0.8137001385896282,
|
| 36 |
+
"eval_loss": 0.10469090193510056,
|
| 37 |
+
"eval_runtime": 35.1587,
|
| 38 |
+
"eval_samples_per_second": 4073.218,
|
| 39 |
+
"eval_steps_per_second": 42.436,
|
| 40 |
+
"step": 5000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.69,
|
| 44 |
+
"learning_rate": 1.9898029417041328e-05,
|
| 45 |
+
"loss": 0.1066,
|
| 46 |
+
"step": 7500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.69,
|
| 50 |
+
"eval_accuracy": 0.8300665460969632,
|
| 51 |
+
"eval_f1": 0.8276076020101577,
|
| 52 |
+
"eval_loss": 0.09453058242797852,
|
| 53 |
+
"eval_runtime": 34.7428,
|
| 54 |
+
"eval_samples_per_second": 4121.976,
|
| 55 |
+
"eval_steps_per_second": 42.944,
|
| 56 |
+
"step": 7500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.25,
|
| 60 |
+
"learning_rate": 1.979570139952044e-05,
|
| 61 |
+
"loss": 0.0975,
|
| 62 |
+
"step": 10000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.25,
|
| 66 |
+
"eval_accuracy": 0.838564615352387,
|
| 67 |
+
"eval_f1": 0.8366911235307668,
|
| 68 |
+
"eval_loss": 0.0888088271021843,
|
| 69 |
+
"eval_runtime": 33.4588,
|
| 70 |
+
"eval_samples_per_second": 4280.16,
|
| 71 |
+
"eval_steps_per_second": 44.592,
|
| 72 |
+
"step": 10000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.81,
|
| 76 |
+
"learning_rate": 1.965864144044478e-05,
|
| 77 |
+
"loss": 0.0917,
|
| 78 |
+
"step": 12500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.81,
|
| 82 |
+
"eval_accuracy": 0.844548876118121,
|
| 83 |
+
"eval_f1": 0.8428155480829799,
|
| 84 |
+
"eval_loss": 0.08494799584150314,
|
| 85 |
+
"eval_runtime": 33.2026,
|
| 86 |
+
"eval_samples_per_second": 4313.187,
|
| 87 |
+
"eval_steps_per_second": 44.936,
|
| 88 |
+
"step": 12500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.37,
|
| 92 |
+
"learning_rate": 1.9487335503840186e-05,
|
| 93 |
+
"loss": 0.0865,
|
| 94 |
+
"step": 15000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 3.37,
|
| 98 |
+
"eval_accuracy": 0.8495625274947803,
|
| 99 |
+
"eval_f1": 0.8483980426142682,
|
| 100 |
+
"eval_loss": 0.08176358044147491,
|
| 101 |
+
"eval_runtime": 34.1293,
|
| 102 |
+
"eval_samples_per_second": 4196.068,
|
| 103 |
+
"eval_steps_per_second": 43.716,
|
| 104 |
+
"step": 15000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.94,
|
| 108 |
+
"learning_rate": 1.9282390977328586e-05,
|
| 109 |
+
"loss": 0.0835,
|
| 110 |
+
"step": 17500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.94,
|
| 114 |
+
"eval_accuracy": 0.852572114881048,
|
| 115 |
+
"eval_f1": 0.8508615234241726,
|
| 116 |
+
"eval_loss": 0.07959919422864914,
|
| 117 |
+
"eval_runtime": 33.0925,
|
| 118 |
+
"eval_samples_per_second": 4327.54,
|
| 119 |
+
"eval_steps_per_second": 45.086,
|
| 120 |
+
"step": 17500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 4.5,
|
| 124 |
+
"learning_rate": 1.904453451855566e-05,
|
| 125 |
+
"loss": 0.08,
|
| 126 |
+
"step": 20000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 4.5,
|
| 130 |
+
"eval_accuracy": 0.8552255793979429,
|
| 131 |
+
"eval_f1": 0.8541677666238712,
|
| 132 |
+
"eval_loss": 0.07768898457288742,
|
| 133 |
+
"eval_runtime": 35.7876,
|
| 134 |
+
"eval_samples_per_second": 4001.64,
|
| 135 |
+
"eval_steps_per_second": 41.69,
|
| 136 |
+
"step": 20000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 5.06,
|
| 140 |
+
"learning_rate": 1.8774609478731048e-05,
|
| 141 |
+
"loss": 0.0778,
|
| 142 |
+
"step": 22500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 5.06,
|
| 146 |
+
"eval_accuracy": 0.8580117171406825,
|
| 147 |
+
"eval_f1": 0.856711042073328,
|
| 148 |
+
"eval_loss": 0.0763072520494461,
|
| 149 |
+
"eval_runtime": 35.4902,
|
| 150 |
+
"eval_samples_per_second": 4035.173,
|
| 151 |
+
"eval_steps_per_second": 42.04,
|
| 152 |
+
"step": 22500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.62,
|
| 156 |
+
"learning_rate": 1.8473572912416232e-05,
|
| 157 |
+
"loss": 0.0753,
|
| 158 |
+
"step": 25000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 5.62,
|
| 162 |
+
"eval_accuracy": 0.8603509555963661,
|
| 163 |
+
"eval_f1": 0.859160334506478,
|
| 164 |
+
"eval_loss": 0.07438770681619644,
|
| 165 |
+
"eval_runtime": 35.3837,
|
| 166 |
+
"eval_samples_per_second": 4047.317,
|
| 167 |
+
"eval_steps_per_second": 42.166,
|
| 168 |
+
"step": 25000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.19,
|
| 172 |
+
"learning_rate": 1.8142492184162323e-05,
|
| 173 |
+
"loss": 0.0739,
|
| 174 |
+
"step": 27500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 6.19,
|
| 178 |
+
"eval_accuracy": 0.8613844101976831,
|
| 179 |
+
"eval_f1": 0.8603226866244833,
|
| 180 |
+
"eval_loss": 0.07380488514900208,
|
| 181 |
+
"eval_runtime": 33.7385,
|
| 182 |
+
"eval_samples_per_second": 4244.676,
|
| 183 |
+
"eval_steps_per_second": 44.222,
|
| 184 |
+
"step": 27500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 6.75,
|
| 188 |
+
"learning_rate": 1.7782541184029316e-05,
|
| 189 |
+
"loss": 0.0716,
|
| 190 |
+
"step": 30000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.75,
|
| 194 |
+
"eval_accuracy": 0.863032351318702,
|
| 195 |
+
"eval_f1": 0.8620277221500123,
|
| 196 |
+
"eval_loss": 0.0729290321469307,
|
| 197 |
+
"eval_runtime": 33.5267,
|
| 198 |
+
"eval_samples_per_second": 4271.486,
|
| 199 |
+
"eval_steps_per_second": 44.502,
|
| 200 |
+
"step": 30000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 7.31,
|
| 204 |
+
"learning_rate": 1.7394996165405244e-05,
|
| 205 |
+
"loss": 0.0701,
|
| 206 |
+
"step": 32500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.31,
|
| 210 |
+
"eval_accuracy": 0.8645196880084353,
|
| 211 |
+
"eval_f1": 0.8638326806071638,
|
| 212 |
+
"eval_loss": 0.07191809266805649,
|
| 213 |
+
"eval_runtime": 35.0686,
|
| 214 |
+
"eval_samples_per_second": 4083.685,
|
| 215 |
+
"eval_steps_per_second": 42.545,
|
| 216 |
+
"step": 32500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 7.87,
|
| 220 |
+
"learning_rate": 1.69812312198827e-05,
|
| 221 |
+
"loss": 0.0689,
|
| 222 |
+
"step": 35000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 7.87,
|
| 226 |
+
"eval_accuracy": 0.8656578846301559,
|
| 227 |
+
"eval_f1": 0.8646806211498433,
|
| 228 |
+
"eval_loss": 0.07080969214439392,
|
| 229 |
+
"eval_runtime": 33.9508,
|
| 230 |
+
"eval_samples_per_second": 4218.132,
|
| 231 |
+
"eval_steps_per_second": 43.946,
|
| 232 |
+
"step": 35000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 8.43,
|
| 236 |
+
"learning_rate": 1.6542713405237254e-05,
|
| 237 |
+
"loss": 0.067,
|
| 238 |
+
"step": 37500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 8.43,
|
| 242 |
+
"eval_accuracy": 0.8670823761076468,
|
| 243 |
+
"eval_f1": 0.8660045143213903,
|
| 244 |
+
"eval_loss": 0.07053036987781525,
|
| 245 |
+
"eval_runtime": 33.5221,
|
| 246 |
+
"eval_samples_per_second": 4272.077,
|
| 247 |
+
"eval_steps_per_second": 44.508,
|
| 248 |
+
"step": 37500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 9.0,
|
| 252 |
+
"learning_rate": 1.6080997543782063e-05,
|
| 253 |
+
"loss": 0.0669,
|
| 254 |
+
"step": 40000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 9.0,
|
| 258 |
+
"eval_accuracy": 0.8681158307089638,
|
| 259 |
+
"eval_f1": 0.8673702293109924,
|
| 260 |
+
"eval_loss": 0.06987718492746353,
|
| 261 |
+
"eval_runtime": 33.3948,
|
| 262 |
+
"eval_samples_per_second": 4288.361,
|
| 263 |
+
"eval_steps_per_second": 44.678,
|
| 264 |
+
"step": 40000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 9.56,
|
| 268 |
+
"learning_rate": 1.5597720709541834e-05,
|
| 269 |
+
"loss": 0.0647,
|
| 270 |
+
"step": 42500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.56,
|
| 274 |
+
"eval_accuracy": 0.8683392803524919,
|
| 275 |
+
"eval_f1": 0.8673043954616042,
|
| 276 |
+
"eval_loss": 0.06969144195318222,
|
| 277 |
+
"eval_runtime": 33.3931,
|
| 278 |
+
"eval_samples_per_second": 4288.583,
|
| 279 |
+
"eval_steps_per_second": 44.68,
|
| 280 |
+
"step": 42500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 10.12,
|
| 284 |
+
"learning_rate": 1.509459642379259e-05,
|
| 285 |
+
"loss": 0.0641,
|
| 286 |
+
"step": 45000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 10.12,
|
| 290 |
+
"eval_accuracy": 0.8690515260912373,
|
| 291 |
+
"eval_f1": 0.8680669250266475,
|
| 292 |
+
"eval_loss": 0.06932760030031204,
|
| 293 |
+
"eval_runtime": 33.6285,
|
| 294 |
+
"eval_samples_per_second": 4258.561,
|
| 295 |
+
"eval_steps_per_second": 44.367,
|
| 296 |
+
"step": 45000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 10.68,
|
| 300 |
+
"learning_rate": 1.4573408579547676e-05,
|
| 301 |
+
"loss": 0.063,
|
| 302 |
+
"step": 47500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 10.68,
|
| 306 |
+
"eval_accuracy": 0.8701827399115978,
|
| 307 |
+
"eval_f1": 0.8693810704654683,
|
| 308 |
+
"eval_loss": 0.06850136071443558,
|
| 309 |
+
"eval_runtime": 33.1134,
|
| 310 |
+
"eval_samples_per_second": 4324.805,
|
| 311 |
+
"eval_steps_per_second": 45.057,
|
| 312 |
+
"step": 47500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 11.25,
|
| 316 |
+
"learning_rate": 1.4036005116531579e-05,
|
| 317 |
+
"loss": 0.0618,
|
| 318 |
+
"step": 50000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 11.25,
|
| 322 |
+
"eval_accuracy": 0.8709089512530637,
|
| 323 |
+
"eval_f1": 0.8700675299699858,
|
| 324 |
+
"eval_loss": 0.06813304871320724,
|
| 325 |
+
"eval_runtime": 33.5088,
|
| 326 |
+
"eval_samples_per_second": 4273.778,
|
| 327 |
+
"eval_steps_per_second": 44.526,
|
| 328 |
+
"step": 50000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 11.81,
|
| 332 |
+
"learning_rate": 1.3484291469067736e-05,
|
| 333 |
+
"loss": 0.0614,
|
| 334 |
+
"step": 52500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 11.81,
|
| 338 |
+
"eval_accuracy": 0.8719843026625421,
|
| 339 |
+
"eval_f1": 0.8711736208762333,
|
| 340 |
+
"eval_loss": 0.0674930214881897,
|
| 341 |
+
"eval_runtime": 33.3571,
|
| 342 |
+
"eval_samples_per_second": 4293.204,
|
| 343 |
+
"eval_steps_per_second": 44.728,
|
| 344 |
+
"step": 52500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 12.37,
|
| 348 |
+
"learning_rate": 1.2920223810111731e-05,
|
| 349 |
+
"loss": 0.0601,
|
| 350 |
+
"step": 55000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.37,
|
| 354 |
+
"eval_accuracy": 0.8723543911346354,
|
| 355 |
+
"eval_f1": 0.8713242180501024,
|
| 356 |
+
"eval_loss": 0.06780188530683517,
|
| 357 |
+
"eval_runtime": 33.8602,
|
| 358 |
+
"eval_samples_per_second": 4229.414,
|
| 359 |
+
"eval_steps_per_second": 44.063,
|
| 360 |
+
"step": 55000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 12.93,
|
| 364 |
+
"learning_rate": 1.2345802115384014e-05,
|
| 365 |
+
"loss": 0.0598,
|
| 366 |
+
"step": 57500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 12.93,
|
| 370 |
+
"eval_accuracy": 0.8731853444965051,
|
| 371 |
+
"eval_f1": 0.8725266467114048,
|
| 372 |
+
"eval_loss": 0.06704463809728622,
|
| 373 |
+
"eval_runtime": 34.9394,
|
| 374 |
+
"eval_samples_per_second": 4098.78,
|
| 375 |
+
"eval_steps_per_second": 42.702,
|
| 376 |
+
"step": 57500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 13.5,
|
| 380 |
+
"learning_rate": 1.1763063072194181e-05,
|
| 381 |
+
"loss": 0.0584,
|
| 382 |
+
"step": 60000
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 13.5,
|
| 386 |
+
"eval_accuracy": 0.8731713788937846,
|
| 387 |
+
"eval_f1": 0.8723142764159807,
|
| 388 |
+
"eval_loss": 0.06695400178432465,
|
| 389 |
+
"eval_runtime": 33.3284,
|
| 390 |
+
"eval_samples_per_second": 4296.9,
|
| 391 |
+
"eval_steps_per_second": 44.767,
|
| 392 |
+
"step": 60000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 14.06,
|
| 396 |
+
"learning_rate": 1.1174072858099545e-05,
|
| 397 |
+
"loss": 0.0584,
|
| 398 |
+
"step": 62500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 14.06,
|
| 402 |
+
"eval_accuracy": 0.8740023322556543,
|
| 403 |
+
"eval_f1": 0.8732025404369645,
|
| 404 |
+
"eval_loss": 0.0664532408118248,
|
| 405 |
+
"eval_runtime": 36.3142,
|
| 406 |
+
"eval_samples_per_second": 3943.604,
|
| 407 |
+
"eval_steps_per_second": 41.086,
|
| 408 |
+
"step": 62500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.62,
|
| 412 |
+
"learning_rate": 1.0580919815002126e-05,
|
| 413 |
+
"loss": 0.0572,
|
| 414 |
+
"step": 65000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 14.62,
|
| 418 |
+
"eval_accuracy": 0.8744282831386295,
|
| 419 |
+
"eval_f1": 0.8734337495737768,
|
| 420 |
+
"eval_loss": 0.06649637967348099,
|
| 421 |
+
"eval_runtime": 33.7341,
|
| 422 |
+
"eval_samples_per_second": 4245.227,
|
| 423 |
+
"eval_steps_per_second": 44.228,
|
| 424 |
+
"step": 65000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 15.18,
|
| 428 |
+
"learning_rate": 9.98570704465907e-06,
|
| 429 |
+
"loss": 0.0567,
|
| 430 |
+
"step": 67500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 15.18,
|
| 434 |
+
"eval_accuracy": 0.8752732021032198,
|
| 435 |
+
"eval_f1": 0.8744599630053497,
|
| 436 |
+
"eval_loss": 0.06611284613609314,
|
| 437 |
+
"eval_runtime": 33.3726,
|
| 438 |
+
"eval_samples_per_second": 4291.211,
|
| 439 |
+
"eval_steps_per_second": 44.707,
|
| 440 |
+
"step": 67500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 15.74,
|
| 444 |
+
"learning_rate": 9.390544951860105e-06,
|
| 445 |
+
"loss": 0.0561,
|
| 446 |
+
"step": 70000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 15.74,
|
| 450 |
+
"eval_accuracy": 0.8756293249725925,
|
| 451 |
+
"eval_f1": 0.8749748466805782,
|
| 452 |
+
"eval_loss": 0.06604801118373871,
|
| 453 |
+
"eval_runtime": 33.5726,
|
| 454 |
+
"eval_samples_per_second": 4265.656,
|
| 455 |
+
"eval_steps_per_second": 44.441,
|
| 456 |
+
"step": 70000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 16.31,
|
| 460 |
+
"learning_rate": 8.797543761711079e-06,
|
| 461 |
+
"loss": 0.0554,
|
| 462 |
+
"step": 72500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 16.31,
|
| 466 |
+
"eval_accuracy": 0.8758876886229218,
|
| 467 |
+
"eval_f1": 0.8750835415020012,
|
| 468 |
+
"eval_loss": 0.06606367230415344,
|
| 469 |
+
"eval_runtime": 35.9562,
|
| 470 |
+
"eval_samples_per_second": 3982.87,
|
| 471 |
+
"eval_steps_per_second": 41.495,
|
| 472 |
+
"step": 72500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 16.87,
|
| 476 |
+
"learning_rate": 8.208806037554645e-06,
|
| 477 |
+
"loss": 0.0552,
|
| 478 |
+
"step": 75000
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 16.87,
|
| 482 |
+
"eval_accuracy": 0.875503634548108,
|
| 483 |
+
"eval_f1": 0.8748588561400703,
|
| 484 |
+
"eval_loss": 0.06561503559350967,
|
| 485 |
+
"eval_runtime": 33.3706,
|
| 486 |
+
"eval_samples_per_second": 4291.466,
|
| 487 |
+
"eval_steps_per_second": 44.71,
|
| 488 |
+
"step": 75000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 17.43,
|
| 492 |
+
"learning_rate": 7.6264192260566915e-06,
|
| 493 |
+
"loss": 0.0544,
|
| 494 |
+
"step": 77500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 17.43,
|
| 498 |
+
"eval_accuracy": 0.8761670006773318,
|
| 499 |
+
"eval_f1": 0.875430169191633,
|
| 500 |
+
"eval_loss": 0.06571561098098755,
|
| 501 |
+
"eval_runtime": 33.8002,
|
| 502 |
+
"eval_samples_per_second": 4236.932,
|
| 503 |
+
"eval_steps_per_second": 44.142,
|
| 504 |
+
"step": 77500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 17.99,
|
| 508 |
+
"learning_rate": 7.052448255890958e-06,
|
| 509 |
+
"loss": 0.0544,
|
| 510 |
+
"step": 80000
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 17.99,
|
| 514 |
+
"eval_accuracy": 0.8766627795739095,
|
| 515 |
+
"eval_f1": 0.8759974000823453,
|
| 516 |
+
"eval_loss": 0.06542336195707321,
|
| 517 |
+
"eval_runtime": 34.4239,
|
| 518 |
+
"eval_samples_per_second": 4160.157,
|
| 519 |
+
"eval_steps_per_second": 43.342,
|
| 520 |
+
"step": 80000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 18.56,
|
| 524 |
+
"learning_rate": 6.488928216264112e-06,
|
| 525 |
+
"loss": 0.0534,
|
| 526 |
+
"step": 82500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 18.56,
|
| 530 |
+
"eval_accuracy": 0.8767116591834312,
|
| 531 |
+
"eval_f1": 0.8759135326151816,
|
| 532 |
+
"eval_loss": 0.06542443484067917,
|
| 533 |
+
"eval_runtime": 33.2655,
|
| 534 |
+
"eval_samples_per_second": 4305.035,
|
| 535 |
+
"eval_steps_per_second": 44.851,
|
| 536 |
+
"step": 82500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 19.12,
|
| 540 |
+
"learning_rate": 5.93785714124059e-06,
|
| 541 |
+
"loss": 0.0534,
|
| 542 |
+
"step": 85000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 19.12,
|
| 546 |
+
"eval_accuracy": 0.8773470941072139,
|
| 547 |
+
"eval_f1": 0.8766532567238048,
|
| 548 |
+
"eval_loss": 0.06530272215604782,
|
| 549 |
+
"eval_runtime": 35.0124,
|
| 550 |
+
"eval_samples_per_second": 4090.241,
|
| 551 |
+
"eval_steps_per_second": 42.614,
|
| 552 |
+
"step": 85000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 19.68,
|
| 556 |
+
"learning_rate": 5.401188925451274e-06,
|
| 557 |
+
"loss": 0.0528,
|
| 558 |
+
"step": 87500
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 19.68,
|
| 562 |
+
"eval_accuracy": 0.8775146813398599,
|
| 563 |
+
"eval_f1": 0.8768256135826535,
|
| 564 |
+
"eval_loss": 0.06490106880664825,
|
| 565 |
+
"eval_runtime": 33.5689,
|
| 566 |
+
"eval_samples_per_second": 4266.118,
|
| 567 |
+
"eval_steps_per_second": 44.446,
|
| 568 |
+
"step": 87500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 20.24,
|
| 572 |
+
"learning_rate": 4.880826396304312e-06,
|
| 573 |
+
"loss": 0.0525,
|
| 574 |
+
"step": 90000
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 20.24,
|
| 578 |
+
"eval_accuracy": 0.8776264061616239,
|
| 579 |
+
"eval_f1": 0.8768652778836793,
|
| 580 |
+
"eval_loss": 0.06507841497659683,
|
| 581 |
+
"eval_runtime": 33.3247,
|
| 582 |
+
"eval_samples_per_second": 4297.383,
|
| 583 |
+
"eval_steps_per_second": 44.772,
|
| 584 |
+
"step": 90000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 20.81,
|
| 588 |
+
"learning_rate": 4.378614567261487e-06,
|
| 589 |
+
"loss": 0.0523,
|
| 590 |
+
"step": 92500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 20.81,
|
| 594 |
+
"eval_accuracy": 0.8775007157371394,
|
| 595 |
+
"eval_f1": 0.8767952508584296,
|
| 596 |
+
"eval_loss": 0.06489618122577667,
|
| 597 |
+
"eval_runtime": 36.2651,
|
| 598 |
+
"eval_samples_per_second": 3948.949,
|
| 599 |
+
"eval_steps_per_second": 41.141,
|
| 600 |
+
"step": 92500
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 21.37,
|
| 604 |
+
"learning_rate": 3.896334096101447e-06,
|
| 605 |
+
"loss": 0.0517,
|
| 606 |
+
"step": 95000
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 21.37,
|
| 610 |
+
"eval_accuracy": 0.8782339098799656,
|
| 611 |
+
"eval_f1": 0.877492038187804,
|
| 612 |
+
"eval_loss": 0.06479762494564056,
|
| 613 |
+
"eval_runtime": 33.4015,
|
| 614 |
+
"eval_samples_per_second": 4287.5,
|
| 615 |
+
"eval_steps_per_second": 44.669,
|
| 616 |
+
"step": 95000
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 21.93,
|
| 620 |
+
"learning_rate": 3.4356949713644915e-06,
|
| 621 |
+
"loss": 0.0516,
|
| 622 |
+
"step": 97500
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 21.93,
|
| 626 |
+
"eval_accuracy": 0.8782897722908476,
|
| 627 |
+
"eval_f1": 0.8775965653086075,
|
| 628 |
+
"eval_loss": 0.06478870660066605,
|
| 629 |
+
"eval_runtime": 33.7278,
|
| 630 |
+
"eval_samples_per_second": 4246.021,
|
| 631 |
+
"eval_steps_per_second": 44.236,
|
| 632 |
+
"step": 97500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 22.49,
|
| 636 |
+
"learning_rate": 2.9983304493643495e-06,
|
| 637 |
+
"loss": 0.0511,
|
| 638 |
+
"step": 100000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 22.49,
|
| 642 |
+
"eval_accuracy": 0.8780523570445992,
|
| 643 |
+
"eval_f1": 0.8773587469443522,
|
| 644 |
+
"eval_loss": 0.06483691185712814,
|
| 645 |
+
"eval_runtime": 33.3746,
|
| 646 |
+
"eval_samples_per_second": 4290.959,
|
| 647 |
+
"eval_steps_per_second": 44.705,
|
| 648 |
+
"step": 100000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 23.05,
|
| 652 |
+
"learning_rate": 2.5857912632641447e-06,
|
| 653 |
+
"loss": 0.0511,
|
| 654 |
+
"step": 102500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 23.05,
|
| 658 |
+
"eval_accuracy": 0.8783246862976489,
|
| 659 |
+
"eval_f1": 0.8776168302082048,
|
| 660 |
+
"eval_loss": 0.06472069770097733,
|
| 661 |
+
"eval_runtime": 33.1429,
|
| 662 |
+
"eval_samples_per_second": 4320.953,
|
| 663 |
+
"eval_steps_per_second": 45.017,
|
| 664 |
+
"step": 102500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 23.62,
|
| 668 |
+
"learning_rate": 2.199540124748957e-06,
|
| 669 |
+
"loss": 0.0508,
|
| 670 |
+
"step": 105000
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 23.62,
|
| 674 |
+
"eval_accuracy": 0.8785062391330154,
|
| 675 |
+
"eval_f1": 0.8777938106114025,
|
| 676 |
+
"eval_loss": 0.06474106758832932,
|
| 677 |
+
"eval_runtime": 33.3508,
|
| 678 |
+
"eval_samples_per_second": 4294.023,
|
| 679 |
+
"eval_steps_per_second": 44.737,
|
| 680 |
+
"step": 105000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 24.18,
|
| 684 |
+
"learning_rate": 1.8409465377900981e-06,
|
| 685 |
+
"loss": 0.0505,
|
| 686 |
+
"step": 107500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 24.18,
|
| 690 |
+
"eval_accuracy": 0.8785341703384564,
|
| 691 |
+
"eval_f1": 0.8777444031001141,
|
| 692 |
+
"eval_loss": 0.0646664947271347,
|
| 693 |
+
"eval_runtime": 33.8974,
|
| 694 |
+
"eval_samples_per_second": 4224.776,
|
| 695 |
+
"eval_steps_per_second": 44.015,
|
| 696 |
+
"step": 107500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 24.74,
|
| 700 |
+
"learning_rate": 1.5112819428894976e-06,
|
| 701 |
+
"loss": 0.0505,
|
| 702 |
+
"step": 110000
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 24.74,
|
| 706 |
+
"eval_accuracy": 0.8787855511874254,
|
| 707 |
+
"eval_f1": 0.8780881956947786,
|
| 708 |
+
"eval_loss": 0.06458932906389236,
|
| 709 |
+
"eval_runtime": 33.1686,
|
| 710 |
+
"eval_samples_per_second": 4317.613,
|
| 711 |
+
"eval_steps_per_second": 44.982,
|
| 712 |
+
"step": 110000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 25.3,
|
| 716 |
+
"learning_rate": 1.2117152090209806e-06,
|
| 717 |
+
"loss": 0.0503,
|
| 718 |
+
"step": 112500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 25.3,
|
| 722 |
+
"eval_accuracy": 0.8786389123588602,
|
| 723 |
+
"eval_f1": 0.8779011228018737,
|
| 724 |
+
"eval_loss": 0.06458309292793274,
|
| 725 |
+
"eval_runtime": 33.1934,
|
| 726 |
+
"eval_samples_per_second": 4314.381,
|
| 727 |
+
"eval_steps_per_second": 44.949,
|
| 728 |
+
"step": 112500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 25.87,
|
| 732 |
+
"learning_rate": 9.433084892523181e-07,
|
| 733 |
+
"loss": 0.0502,
|
| 734 |
+
"step": 115000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 25.87,
|
| 738 |
+
"eval_accuracy": 0.8788623620023881,
|
| 739 |
+
"eval_f1": 0.878161188935205,
|
| 740 |
+
"eval_loss": 0.06459838151931763,
|
| 741 |
+
"eval_runtime": 35.5543,
|
| 742 |
+
"eval_samples_per_second": 4027.899,
|
| 743 |
+
"eval_steps_per_second": 41.964,
|
| 744 |
+
"step": 115000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 26.43,
|
| 748 |
+
"learning_rate": 7.070134547424945e-07,
|
| 749 |
+
"loss": 0.0501,
|
| 750 |
+
"step": 117500
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 26.43,
|
| 754 |
+
"eval_accuracy": 0.8787995167901459,
|
| 755 |
+
"eval_f1": 0.8780747579631859,
|
| 756 |
+
"eval_loss": 0.06461162865161896,
|
| 757 |
+
"eval_runtime": 33.4871,
|
| 758 |
+
"eval_samples_per_second": 4276.544,
|
| 759 |
+
"eval_steps_per_second": 44.554,
|
| 760 |
+
"step": 117500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 26.99,
|
| 764 |
+
"learning_rate": 5.036679204670313e-07,
|
| 765 |
+
"loss": 0.0501,
|
| 766 |
+
"step": 120000
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 26.99,
|
| 770 |
+
"eval_accuracy": 0.8791067600499969,
|
| 771 |
+
"eval_f1": 0.8784083866618659,
|
| 772 |
+
"eval_loss": 0.06452779471874237,
|
| 773 |
+
"eval_runtime": 33.2577,
|
| 774 |
+
"eval_samples_per_second": 4306.037,
|
| 775 |
+
"eval_steps_per_second": 44.862,
|
| 776 |
+
"step": 120000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 27.55,
|
| 780 |
+
"learning_rate": 3.339928746353327e-07,
|
| 781 |
+
"loss": 0.05,
|
| 782 |
+
"step": 122500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 27.55,
|
| 786 |
+
"eval_accuracy": 0.8790439148377546,
|
| 787 |
+
"eval_f1": 0.8783328357475545,
|
| 788 |
+
"eval_loss": 0.06455225497484207,
|
| 789 |
+
"eval_runtime": 33.5082,
|
| 790 |
+
"eval_samples_per_second": 4273.851,
|
| 791 |
+
"eval_steps_per_second": 44.526,
|
| 792 |
+
"step": 122500
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 28.12,
|
| 796 |
+
"learning_rate": 1.9858992233260598e-07,
|
| 797 |
+
"loss": 0.0497,
|
| 798 |
+
"step": 125000
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 28.12,
|
| 802 |
+
"eval_accuracy": 0.8791626224608788,
|
| 803 |
+
"eval_f1": 0.8784598679092687,
|
| 804 |
+
"eval_loss": 0.06452032178640366,
|
| 805 |
+
"eval_runtime": 33.4248,
|
| 806 |
+
"eval_samples_per_second": 4284.51,
|
| 807 |
+
"eval_steps_per_second": 44.637,
|
| 808 |
+
"step": 125000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 28.68,
|
| 812 |
+
"learning_rate": 9.793915245028595e-08,
|
| 813 |
+
"loss": 0.0499,
|
| 814 |
+
"step": 127500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 28.68,
|
| 818 |
+
"eval_accuracy": 0.8791277084540776,
|
| 819 |
+
"eval_f1": 0.8784333454818922,
|
| 820 |
+
"eval_loss": 0.06449923664331436,
|
| 821 |
+
"eval_runtime": 35.6969,
|
| 822 |
+
"eval_samples_per_second": 4011.8,
|
| 823 |
+
"eval_steps_per_second": 41.796,
|
| 824 |
+
"step": 127500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 29.24,
|
| 828 |
+
"learning_rate": 3.239743546802565e-08,
|
| 829 |
+
"loss": 0.0499,
|
| 830 |
+
"step": 130000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 29.24,
|
| 834 |
+
"eval_accuracy": 0.8791556396595186,
|
| 835 |
+
"eval_f1": 0.8784619602482451,
|
| 836 |
+
"eval_loss": 0.06451133638620377,
|
| 837 |
+
"eval_runtime": 33.3523,
|
| 838 |
+
"eval_samples_per_second": 4293.833,
|
| 839 |
+
"eval_steps_per_second": 44.735,
|
| 840 |
+
"step": 130000
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 29.8,
|
| 844 |
+
"learning_rate": 2.197158122699827e-09,
|
| 845 |
+
"loss": 0.0497,
|
| 846 |
+
"step": 132500
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 29.8,
|
| 850 |
+
"eval_accuracy": 0.8791137428513571,
|
| 851 |
+
"eval_f1": 0.8784195035720077,
|
| 852 |
+
"eval_loss": 0.06450776755809784,
|
| 853 |
+
"eval_runtime": 33.5967,
|
| 854 |
+
"eval_samples_per_second": 4262.595,
|
| 855 |
+
"eval_steps_per_second": 44.409,
|
| 856 |
+
"step": 132500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 30.0,
|
| 860 |
+
"step": 133380,
|
| 861 |
+
"total_flos": 1.6627855289686797e+17,
|
| 862 |
+
"train_loss": 0.06639543622843883,
|
| 863 |
+
"train_runtime": 19911.3581,
|
| 864 |
+
"train_samples_per_second": 1714.776,
|
| 865 |
+
"train_steps_per_second": 6.699
|
| 866 |
+
}
|
| 867 |
+
],
|
| 868 |
+
"max_steps": 133380,
|
| 869 |
+
"num_train_epochs": 30,
|
| 870 |
+
"total_flos": 1.6627855289686797e+17,
|
| 871 |
+
"trial_name": null,
|
| 872 |
+
"trial_params": null
|
| 873 |
+
}
|