🍻 cheers
Browse files- README.md +6 -5
- all_results.json +14 -0
- eval_results.json +10 -0
- runs/Mar31_13-20-20_ryanserver/events.out.tfevents.1711919481.ryanserver.4019.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1643 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
model-index:
|
| 7 |
- name: ryan03312024_lr_2e-5_wd_001
|
|
@@ -13,12 +14,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# ryan03312024_lr_2e-5_wd_001
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
-
- Ordinal Mae: 0.
|
| 20 |
-
- Ordinal Accuracy: 0.
|
| 21 |
-
- Na Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: ryan03312024_lr_2e-5_wd_001
|
|
|
|
| 14 |
|
| 15 |
# ryan03312024_lr_2e-5_wd_001
|
| 16 |
|
| 17 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the properties dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.1916
|
| 20 |
+
- Ordinal Mae: 0.4221
|
| 21 |
+
- Ordinal Accuracy: 0.6828
|
| 22 |
+
- Na Accuracy: 0.8591
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.5,
|
| 3 |
+
"eval_loss": 0.19156721234321594,
|
| 4 |
+
"eval_na_accuracy": 0.8591065406799316,
|
| 5 |
+
"eval_ordinal_accuracy": 0.682763934135437,
|
| 6 |
+
"eval_ordinal_mae": 0.42213648557662964,
|
| 7 |
+
"eval_runtime": 192.9438,
|
| 8 |
+
"eval_samples_per_second": 23.193,
|
| 9 |
+
"eval_steps_per_second": 2.902,
|
| 10 |
+
"train_loss": 0.21532289660893955,
|
| 11 |
+
"train_runtime": 13643.9574,
|
| 12 |
+
"train_samples_per_second": 4.878,
|
| 13 |
+
"train_steps_per_second": 0.305
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.5,
|
| 3 |
+
"eval_loss": 0.19156721234321594,
|
| 4 |
+
"eval_na_accuracy": 0.8591065406799316,
|
| 5 |
+
"eval_ordinal_accuracy": 0.682763934135437,
|
| 6 |
+
"eval_ordinal_mae": 0.42213648557662964,
|
| 7 |
+
"eval_runtime": 192.9438,
|
| 8 |
+
"eval_samples_per_second": 23.193,
|
| 9 |
+
"eval_steps_per_second": 2.902
|
| 10 |
+
}
|
runs/Mar31_13-20-20_ryanserver/events.out.tfevents.1711919481.ryanserver.4019.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d39b179c0651cc44bd5a623e1ac5c555f90ce58b4c437dab7bdb40cf4c2ea88d
|
| 3 |
+
size 529
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.5,
|
| 3 |
+
"train_loss": 0.21532289660893955,
|
| 4 |
+
"train_runtime": 13643.9574,
|
| 5 |
+
"train_samples_per_second": 4.878,
|
| 6 |
+
"train_steps_per_second": 0.305
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1643 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.19156721234321594,
|
| 3 |
+
"best_model_checkpoint": "./ryan03312024_lr_2e-5_wd_001/checkpoint-3100",
|
| 4 |
+
"epoch": 1.5001803101334295,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 4160,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.6417293548583984,
|
| 14 |
+
"learning_rate": 1.9879807692307693e-05,
|
| 15 |
+
"loss": 0.5741,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 1.4069948196411133,
|
| 21 |
+
"learning_rate": 1.975961538461539e-05,
|
| 22 |
+
"loss": 0.4638,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03,
|
| 27 |
+
"grad_norm": 1.5593534708023071,
|
| 28 |
+
"learning_rate": 1.963942307692308e-05,
|
| 29 |
+
"loss": 0.4114,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04,
|
| 34 |
+
"grad_norm": 0.6675819158554077,
|
| 35 |
+
"learning_rate": 1.9519230769230772e-05,
|
| 36 |
+
"loss": 0.4436,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04,
|
| 41 |
+
"eval_loss": 0.3697698712348938,
|
| 42 |
+
"eval_na_accuracy": 0.7989690899848938,
|
| 43 |
+
"eval_ordinal_accuracy": 0.3331620991230011,
|
| 44 |
+
"eval_ordinal_mae": 0.8705630302429199,
|
| 45 |
+
"eval_runtime": 346.2992,
|
| 46 |
+
"eval_samples_per_second": 12.922,
|
| 47 |
+
"eval_steps_per_second": 1.617,
|
| 48 |
+
"step": 100
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.05,
|
| 52 |
+
"grad_norm": 0.6941640377044678,
|
| 53 |
+
"learning_rate": 1.9399038461538464e-05,
|
| 54 |
+
"loss": 0.3901,
|
| 55 |
+
"step": 125
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.05,
|
| 59 |
+
"grad_norm": 0.7081687450408936,
|
| 60 |
+
"learning_rate": 1.9278846153846155e-05,
|
| 61 |
+
"loss": 0.366,
|
| 62 |
+
"step": 150
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.06,
|
| 66 |
+
"grad_norm": 0.9151293635368347,
|
| 67 |
+
"learning_rate": 1.9158653846153847e-05,
|
| 68 |
+
"loss": 0.3773,
|
| 69 |
+
"step": 175
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.07,
|
| 73 |
+
"grad_norm": 0.5204830169677734,
|
| 74 |
+
"learning_rate": 1.903846153846154e-05,
|
| 75 |
+
"loss": 0.3143,
|
| 76 |
+
"step": 200
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.07,
|
| 80 |
+
"eval_loss": 0.3215162754058838,
|
| 81 |
+
"eval_na_accuracy": 0.8092783689498901,
|
| 82 |
+
"eval_ordinal_accuracy": 0.4017467200756073,
|
| 83 |
+
"eval_ordinal_mae": 0.855476975440979,
|
| 84 |
+
"eval_runtime": 209.7588,
|
| 85 |
+
"eval_samples_per_second": 21.334,
|
| 86 |
+
"eval_steps_per_second": 2.67,
|
| 87 |
+
"step": 200
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.08,
|
| 91 |
+
"grad_norm": 1.357649803161621,
|
| 92 |
+
"learning_rate": 1.8918269230769234e-05,
|
| 93 |
+
"loss": 0.3559,
|
| 94 |
+
"step": 225
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.09,
|
| 98 |
+
"grad_norm": 0.7635074257850647,
|
| 99 |
+
"learning_rate": 1.8798076923076926e-05,
|
| 100 |
+
"loss": 0.3616,
|
| 101 |
+
"step": 250
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.1,
|
| 105 |
+
"grad_norm": 2.4480478763580322,
|
| 106 |
+
"learning_rate": 1.8677884615384617e-05,
|
| 107 |
+
"loss": 0.3209,
|
| 108 |
+
"step": 275
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11,
|
| 112 |
+
"grad_norm": 0.8215653300285339,
|
| 113 |
+
"learning_rate": 1.855769230769231e-05,
|
| 114 |
+
"loss": 0.3385,
|
| 115 |
+
"step": 300
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.11,
|
| 119 |
+
"eval_loss": 0.2996984124183655,
|
| 120 |
+
"eval_na_accuracy": 0.8591065406799316,
|
| 121 |
+
"eval_ordinal_accuracy": 0.44849729537963867,
|
| 122 |
+
"eval_ordinal_mae": 0.8302922248840332,
|
| 123 |
+
"eval_runtime": 193.7095,
|
| 124 |
+
"eval_samples_per_second": 23.102,
|
| 125 |
+
"eval_steps_per_second": 2.891,
|
| 126 |
+
"step": 300
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.12,
|
| 130 |
+
"grad_norm": 0.48835399746894836,
|
| 131 |
+
"learning_rate": 1.84375e-05,
|
| 132 |
+
"loss": 0.2976,
|
| 133 |
+
"step": 325
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.13,
|
| 137 |
+
"grad_norm": 0.6756967306137085,
|
| 138 |
+
"learning_rate": 1.8317307692307693e-05,
|
| 139 |
+
"loss": 0.3043,
|
| 140 |
+
"step": 350
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.14,
|
| 144 |
+
"grad_norm": 0.20951023697853088,
|
| 145 |
+
"learning_rate": 1.8197115384615388e-05,
|
| 146 |
+
"loss": 0.3046,
|
| 147 |
+
"step": 375
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.14,
|
| 151 |
+
"grad_norm": 0.9027990102767944,
|
| 152 |
+
"learning_rate": 1.807692307692308e-05,
|
| 153 |
+
"loss": 0.3127,
|
| 154 |
+
"step": 400
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.14,
|
| 158 |
+
"eval_loss": 0.28894639015197754,
|
| 159 |
+
"eval_na_accuracy": 0.8745704293251038,
|
| 160 |
+
"eval_ordinal_accuracy": 0.4880554974079132,
|
| 161 |
+
"eval_ordinal_mae": 0.8012504577636719,
|
| 162 |
+
"eval_runtime": 195.2283,
|
| 163 |
+
"eval_samples_per_second": 22.922,
|
| 164 |
+
"eval_steps_per_second": 2.868,
|
| 165 |
+
"step": 400
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.15,
|
| 169 |
+
"grad_norm": 1.5711051225662231,
|
| 170 |
+
"learning_rate": 1.795673076923077e-05,
|
| 171 |
+
"loss": 0.3094,
|
| 172 |
+
"step": 425
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.16,
|
| 176 |
+
"grad_norm": 2.6442267894744873,
|
| 177 |
+
"learning_rate": 1.7836538461538463e-05,
|
| 178 |
+
"loss": 0.2979,
|
| 179 |
+
"step": 450
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.17,
|
| 183 |
+
"grad_norm": 1.7498008012771606,
|
| 184 |
+
"learning_rate": 1.7716346153846155e-05,
|
| 185 |
+
"loss": 0.3193,
|
| 186 |
+
"step": 475
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.18,
|
| 190 |
+
"grad_norm": 1.4141407012939453,
|
| 191 |
+
"learning_rate": 1.7596153846153846e-05,
|
| 192 |
+
"loss": 0.3054,
|
| 193 |
+
"step": 500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.18,
|
| 197 |
+
"eval_loss": 0.28038087487220764,
|
| 198 |
+
"eval_na_accuracy": 0.8780068755149841,
|
| 199 |
+
"eval_ordinal_accuracy": 0.5324942469596863,
|
| 200 |
+
"eval_ordinal_mae": 0.7619425058364868,
|
| 201 |
+
"eval_runtime": 189.7487,
|
| 202 |
+
"eval_samples_per_second": 23.584,
|
| 203 |
+
"eval_steps_per_second": 2.951,
|
| 204 |
+
"step": 500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.19,
|
| 208 |
+
"grad_norm": 1.3073471784591675,
|
| 209 |
+
"learning_rate": 1.7475961538461538e-05,
|
| 210 |
+
"loss": 0.302,
|
| 211 |
+
"step": 525
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.2,
|
| 215 |
+
"grad_norm": 1.1622358560562134,
|
| 216 |
+
"learning_rate": 1.7355769230769233e-05,
|
| 217 |
+
"loss": 0.2844,
|
| 218 |
+
"step": 550
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.21,
|
| 222 |
+
"grad_norm": 0.3688335716724396,
|
| 223 |
+
"learning_rate": 1.7235576923076925e-05,
|
| 224 |
+
"loss": 0.2745,
|
| 225 |
+
"step": 575
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.22,
|
| 229 |
+
"grad_norm": 1.1387437582015991,
|
| 230 |
+
"learning_rate": 1.7115384615384617e-05,
|
| 231 |
+
"loss": 0.3051,
|
| 232 |
+
"step": 600
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.22,
|
| 236 |
+
"eval_loss": 0.27521631121635437,
|
| 237 |
+
"eval_na_accuracy": 0.9158075451850891,
|
| 238 |
+
"eval_ordinal_accuracy": 0.5235037207603455,
|
| 239 |
+
"eval_ordinal_mae": 0.7215057015419006,
|
| 240 |
+
"eval_runtime": 192.1403,
|
| 241 |
+
"eval_samples_per_second": 23.29,
|
| 242 |
+
"eval_steps_per_second": 2.915,
|
| 243 |
+
"step": 600
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.23,
|
| 247 |
+
"grad_norm": 0.613787055015564,
|
| 248 |
+
"learning_rate": 1.699519230769231e-05,
|
| 249 |
+
"loss": 0.3017,
|
| 250 |
+
"step": 625
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.23,
|
| 254 |
+
"grad_norm": 0.595897376537323,
|
| 255 |
+
"learning_rate": 1.6875e-05,
|
| 256 |
+
"loss": 0.2806,
|
| 257 |
+
"step": 650
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.24,
|
| 261 |
+
"grad_norm": 1.4324086904525757,
|
| 262 |
+
"learning_rate": 1.6754807692307692e-05,
|
| 263 |
+
"loss": 0.256,
|
| 264 |
+
"step": 675
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.25,
|
| 268 |
+
"grad_norm": 3.7929930686950684,
|
| 269 |
+
"learning_rate": 1.6634615384615387e-05,
|
| 270 |
+
"loss": 0.2833,
|
| 271 |
+
"step": 700
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.25,
|
| 275 |
+
"eval_loss": 0.26531103253364563,
|
| 276 |
+
"eval_na_accuracy": 0.8969072103500366,
|
| 277 |
+
"eval_ordinal_accuracy": 0.5486770868301392,
|
| 278 |
+
"eval_ordinal_mae": 0.6806999444961548,
|
| 279 |
+
"eval_runtime": 191.9358,
|
| 280 |
+
"eval_samples_per_second": 23.315,
|
| 281 |
+
"eval_steps_per_second": 2.918,
|
| 282 |
+
"step": 700
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.26,
|
| 286 |
+
"grad_norm": 0.27854958176612854,
|
| 287 |
+
"learning_rate": 1.651442307692308e-05,
|
| 288 |
+
"loss": 0.2805,
|
| 289 |
+
"step": 725
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.27,
|
| 293 |
+
"grad_norm": 1.8652201890945435,
|
| 294 |
+
"learning_rate": 1.6399038461538462e-05,
|
| 295 |
+
"loss": 0.2737,
|
| 296 |
+
"step": 750
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.28,
|
| 300 |
+
"grad_norm": 2.1118507385253906,
|
| 301 |
+
"learning_rate": 1.6278846153846154e-05,
|
| 302 |
+
"loss": 0.2611,
|
| 303 |
+
"step": 775
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.29,
|
| 307 |
+
"grad_norm": 2.6029460430145264,
|
| 308 |
+
"learning_rate": 1.6158653846153845e-05,
|
| 309 |
+
"loss": 0.2907,
|
| 310 |
+
"step": 800
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.29,
|
| 314 |
+
"eval_loss": 0.255000501871109,
|
| 315 |
+
"eval_na_accuracy": 0.8350515365600586,
|
| 316 |
+
"eval_ordinal_accuracy": 0.5617775321006775,
|
| 317 |
+
"eval_ordinal_mae": 0.6431577205657959,
|
| 318 |
+
"eval_runtime": 190.5444,
|
| 319 |
+
"eval_samples_per_second": 23.485,
|
| 320 |
+
"eval_steps_per_second": 2.939,
|
| 321 |
+
"step": 800
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.3,
|
| 325 |
+
"grad_norm": 1.63986337184906,
|
| 326 |
+
"learning_rate": 1.603846153846154e-05,
|
| 327 |
+
"loss": 0.2661,
|
| 328 |
+
"step": 825
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.31,
|
| 332 |
+
"grad_norm": 0.8340407013893127,
|
| 333 |
+
"learning_rate": 1.5918269230769232e-05,
|
| 334 |
+
"loss": 0.2513,
|
| 335 |
+
"step": 850
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.32,
|
| 339 |
+
"grad_norm": 1.321059226989746,
|
| 340 |
+
"learning_rate": 1.5798076923076924e-05,
|
| 341 |
+
"loss": 0.2676,
|
| 342 |
+
"step": 875
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.32,
|
| 346 |
+
"grad_norm": 2.41912841796875,
|
| 347 |
+
"learning_rate": 1.567788461538462e-05,
|
| 348 |
+
"loss": 0.2468,
|
| 349 |
+
"step": 900
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.32,
|
| 353 |
+
"eval_loss": 0.2521895170211792,
|
| 354 |
+
"eval_na_accuracy": 0.8058419227600098,
|
| 355 |
+
"eval_ordinal_accuracy": 0.5972257852554321,
|
| 356 |
+
"eval_ordinal_mae": 0.6118690967559814,
|
| 357 |
+
"eval_runtime": 197.239,
|
| 358 |
+
"eval_samples_per_second": 22.688,
|
| 359 |
+
"eval_steps_per_second": 2.839,
|
| 360 |
+
"step": 900
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.33,
|
| 364 |
+
"grad_norm": 0.9854594469070435,
|
| 365 |
+
"learning_rate": 1.555769230769231e-05,
|
| 366 |
+
"loss": 0.2727,
|
| 367 |
+
"step": 925
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.34,
|
| 371 |
+
"grad_norm": 5.904122352600098,
|
| 372 |
+
"learning_rate": 1.54375e-05,
|
| 373 |
+
"loss": 0.3027,
|
| 374 |
+
"step": 950
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.35,
|
| 378 |
+
"grad_norm": 1.8180320262908936,
|
| 379 |
+
"learning_rate": 1.5317307692307694e-05,
|
| 380 |
+
"loss": 0.2465,
|
| 381 |
+
"step": 975
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.36,
|
| 385 |
+
"grad_norm": 0.5673078894615173,
|
| 386 |
+
"learning_rate": 1.5197115384615386e-05,
|
| 387 |
+
"loss": 0.2199,
|
| 388 |
+
"step": 1000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.36,
|
| 392 |
+
"eval_loss": 0.24374203383922577,
|
| 393 |
+
"eval_na_accuracy": 0.8127147555351257,
|
| 394 |
+
"eval_ordinal_accuracy": 0.606216311454773,
|
| 395 |
+
"eval_ordinal_mae": 0.6023499965667725,
|
| 396 |
+
"eval_runtime": 192.8009,
|
| 397 |
+
"eval_samples_per_second": 23.21,
|
| 398 |
+
"eval_steps_per_second": 2.905,
|
| 399 |
+
"step": 1000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.37,
|
| 403 |
+
"grad_norm": 0.8171074390411377,
|
| 404 |
+
"learning_rate": 1.5076923076923078e-05,
|
| 405 |
+
"loss": 0.221,
|
| 406 |
+
"step": 1025
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.38,
|
| 410 |
+
"grad_norm": 0.881005048751831,
|
| 411 |
+
"learning_rate": 1.495673076923077e-05,
|
| 412 |
+
"loss": 0.235,
|
| 413 |
+
"step": 1050
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.39,
|
| 417 |
+
"grad_norm": 2.021958351135254,
|
| 418 |
+
"learning_rate": 1.4836538461538463e-05,
|
| 419 |
+
"loss": 0.2841,
|
| 420 |
+
"step": 1075
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.4,
|
| 424 |
+
"grad_norm": 1.8785498142242432,
|
| 425 |
+
"learning_rate": 1.4716346153846155e-05,
|
| 426 |
+
"loss": 0.2219,
|
| 427 |
+
"step": 1100
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.4,
|
| 431 |
+
"eval_loss": 0.23609140515327454,
|
| 432 |
+
"eval_na_accuracy": 0.9037800431251526,
|
| 433 |
+
"eval_ordinal_accuracy": 0.595941424369812,
|
| 434 |
+
"eval_ordinal_mae": 0.5573533177375793,
|
| 435 |
+
"eval_runtime": 193.7176,
|
| 436 |
+
"eval_samples_per_second": 23.101,
|
| 437 |
+
"eval_steps_per_second": 2.891,
|
| 438 |
+
"step": 1100
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.41,
|
| 442 |
+
"grad_norm": 0.6871089935302734,
|
| 443 |
+
"learning_rate": 1.4596153846153846e-05,
|
| 444 |
+
"loss": 0.201,
|
| 445 |
+
"step": 1125
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.41,
|
| 449 |
+
"grad_norm": 0.3621855080127716,
|
| 450 |
+
"learning_rate": 1.447596153846154e-05,
|
| 451 |
+
"loss": 0.2119,
|
| 452 |
+
"step": 1150
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.42,
|
| 456 |
+
"grad_norm": 1.404956340789795,
|
| 457 |
+
"learning_rate": 1.4355769230769232e-05,
|
| 458 |
+
"loss": 0.2566,
|
| 459 |
+
"step": 1175
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.43,
|
| 463 |
+
"grad_norm": 0.8949152231216431,
|
| 464 |
+
"learning_rate": 1.4235576923076923e-05,
|
| 465 |
+
"loss": 0.2071,
|
| 466 |
+
"step": 1200
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.43,
|
| 470 |
+
"eval_loss": 0.23867186903953552,
|
| 471 |
+
"eval_na_accuracy": 0.7714776396751404,
|
| 472 |
+
"eval_ordinal_accuracy": 0.6175186038017273,
|
| 473 |
+
"eval_ordinal_mae": 0.5438615679740906,
|
| 474 |
+
"eval_runtime": 196.3391,
|
| 475 |
+
"eval_samples_per_second": 22.792,
|
| 476 |
+
"eval_steps_per_second": 2.852,
|
| 477 |
+
"step": 1200
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.44,
|
| 481 |
+
"grad_norm": 1.7730196714401245,
|
| 482 |
+
"learning_rate": 1.4115384615384617e-05,
|
| 483 |
+
"loss": 0.2756,
|
| 484 |
+
"step": 1225
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.45,
|
| 488 |
+
"grad_norm": 0.7472477555274963,
|
| 489 |
+
"learning_rate": 1.3995192307692308e-05,
|
| 490 |
+
"loss": 0.2318,
|
| 491 |
+
"step": 1250
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.46,
|
| 495 |
+
"grad_norm": 2.942986249923706,
|
| 496 |
+
"learning_rate": 1.3875e-05,
|
| 497 |
+
"loss": 0.2568,
|
| 498 |
+
"step": 1275
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.47,
|
| 502 |
+
"grad_norm": 0.5726996064186096,
|
| 503 |
+
"learning_rate": 1.3754807692307695e-05,
|
| 504 |
+
"loss": 0.2214,
|
| 505 |
+
"step": 1300
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.47,
|
| 509 |
+
"eval_loss": 0.2340591698884964,
|
| 510 |
+
"eval_na_accuracy": 0.7955326437950134,
|
| 511 |
+
"eval_ordinal_accuracy": 0.6231697797775269,
|
| 512 |
+
"eval_ordinal_mae": 0.5256503224372864,
|
| 513 |
+
"eval_runtime": 193.9402,
|
| 514 |
+
"eval_samples_per_second": 23.074,
|
| 515 |
+
"eval_steps_per_second": 2.887,
|
| 516 |
+
"step": 1300
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.48,
|
| 520 |
+
"grad_norm": 0.94366455078125,
|
| 521 |
+
"learning_rate": 1.3634615384615385e-05,
|
| 522 |
+
"loss": 0.2146,
|
| 523 |
+
"step": 1325
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.49,
|
| 527 |
+
"grad_norm": 3.612720251083374,
|
| 528 |
+
"learning_rate": 1.3514423076923077e-05,
|
| 529 |
+
"loss": 0.1979,
|
| 530 |
+
"step": 1350
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.5,
|
| 534 |
+
"grad_norm": 0.878446638584137,
|
| 535 |
+
"learning_rate": 1.3394230769230769e-05,
|
| 536 |
+
"loss": 0.2233,
|
| 537 |
+
"step": 1375
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.5,
|
| 541 |
+
"grad_norm": 1.1531257629394531,
|
| 542 |
+
"learning_rate": 1.3274038461538464e-05,
|
| 543 |
+
"loss": 0.2627,
|
| 544 |
+
"step": 1400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.5,
|
| 548 |
+
"eval_loss": 0.2315448820590973,
|
| 549 |
+
"eval_na_accuracy": 0.7989690899848938,
|
| 550 |
+
"eval_ordinal_accuracy": 0.6123812198638916,
|
| 551 |
+
"eval_ordinal_mae": 0.5152010917663574,
|
| 552 |
+
"eval_runtime": 196.8075,
|
| 553 |
+
"eval_samples_per_second": 22.738,
|
| 554 |
+
"eval_steps_per_second": 2.845,
|
| 555 |
+
"step": 1400
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.51,
|
| 559 |
+
"grad_norm": 2.2066402435302734,
|
| 560 |
+
"learning_rate": 1.3153846153846156e-05,
|
| 561 |
+
"loss": 0.2415,
|
| 562 |
+
"step": 1425
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.52,
|
| 566 |
+
"grad_norm": 1.4936281442642212,
|
| 567 |
+
"learning_rate": 1.3033653846153846e-05,
|
| 568 |
+
"loss": 0.2536,
|
| 569 |
+
"step": 1450
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.53,
|
| 573 |
+
"grad_norm": 2.112527847290039,
|
| 574 |
+
"learning_rate": 1.291346153846154e-05,
|
| 575 |
+
"loss": 0.2105,
|
| 576 |
+
"step": 1475
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.54,
|
| 580 |
+
"grad_norm": 1.8110361099243164,
|
| 581 |
+
"learning_rate": 1.2793269230769233e-05,
|
| 582 |
+
"loss": 0.2067,
|
| 583 |
+
"step": 1500
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.54,
|
| 587 |
+
"eval_loss": 0.22465108335018158,
|
| 588 |
+
"eval_na_accuracy": 0.8109965920448303,
|
| 589 |
+
"eval_ordinal_accuracy": 0.6396095752716064,
|
| 590 |
+
"eval_ordinal_mae": 0.5025707483291626,
|
| 591 |
+
"eval_runtime": 194.1538,
|
| 592 |
+
"eval_samples_per_second": 23.049,
|
| 593 |
+
"eval_steps_per_second": 2.884,
|
| 594 |
+
"step": 1500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.55,
|
| 598 |
+
"grad_norm": 0.8933520913124084,
|
| 599 |
+
"learning_rate": 1.2673076923076924e-05,
|
| 600 |
+
"loss": 0.1957,
|
| 601 |
+
"step": 1525
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.56,
|
| 605 |
+
"grad_norm": 1.3568251132965088,
|
| 606 |
+
"learning_rate": 1.2557692307692309e-05,
|
| 607 |
+
"loss": 0.2286,
|
| 608 |
+
"step": 1550
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.57,
|
| 612 |
+
"grad_norm": 2.600196123123169,
|
| 613 |
+
"learning_rate": 1.24375e-05,
|
| 614 |
+
"loss": 0.2292,
|
| 615 |
+
"step": 1575
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.58,
|
| 619 |
+
"grad_norm": 1.2017379999160767,
|
| 620 |
+
"learning_rate": 1.2317307692307694e-05,
|
| 621 |
+
"loss": 0.2086,
|
| 622 |
+
"step": 1600
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.58,
|
| 626 |
+
"eval_loss": 0.21920213103294373,
|
| 627 |
+
"eval_na_accuracy": 0.8041236996650696,
|
| 628 |
+
"eval_ordinal_accuracy": 0.6588749289512634,
|
| 629 |
+
"eval_ordinal_mae": 0.49550649523735046,
|
| 630 |
+
"eval_runtime": 189.1959,
|
| 631 |
+
"eval_samples_per_second": 23.653,
|
| 632 |
+
"eval_steps_per_second": 2.96,
|
| 633 |
+
"step": 1600
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.59,
|
| 637 |
+
"grad_norm": 2.4263405799865723,
|
| 638 |
+
"learning_rate": 1.2197115384615386e-05,
|
| 639 |
+
"loss": 0.2384,
|
| 640 |
+
"step": 1625
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.6,
|
| 644 |
+
"grad_norm": 2.1560556888580322,
|
| 645 |
+
"learning_rate": 1.2076923076923078e-05,
|
| 646 |
+
"loss": 0.2557,
|
| 647 |
+
"step": 1650
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6,
|
| 651 |
+
"grad_norm": 0.8151688575744629,
|
| 652 |
+
"learning_rate": 1.1956730769230771e-05,
|
| 653 |
+
"loss": 0.2623,
|
| 654 |
+
"step": 1675
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.61,
|
| 658 |
+
"grad_norm": 2.8995933532714844,
|
| 659 |
+
"learning_rate": 1.1836538461538463e-05,
|
| 660 |
+
"loss": 0.1993,
|
| 661 |
+
"step": 1700
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.61,
|
| 665 |
+
"eval_loss": 0.21818678081035614,
|
| 666 |
+
"eval_na_accuracy": 0.8127147555351257,
|
| 667 |
+
"eval_ordinal_accuracy": 0.6521962285041809,
|
| 668 |
+
"eval_ordinal_mae": 0.47375088930130005,
|
| 669 |
+
"eval_runtime": 196.7859,
|
| 670 |
+
"eval_samples_per_second": 22.74,
|
| 671 |
+
"eval_steps_per_second": 2.846,
|
| 672 |
+
"step": 1700
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.62,
|
| 676 |
+
"grad_norm": 0.6673493385314941,
|
| 677 |
+
"learning_rate": 1.1716346153846155e-05,
|
| 678 |
+
"loss": 0.2627,
|
| 679 |
+
"step": 1725
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 0.63,
|
| 683 |
+
"grad_norm": 4.143210411071777,
|
| 684 |
+
"learning_rate": 1.1596153846153848e-05,
|
| 685 |
+
"loss": 0.2294,
|
| 686 |
+
"step": 1750
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.64,
|
| 690 |
+
"grad_norm": 0.9360339641571045,
|
| 691 |
+
"learning_rate": 1.147596153846154e-05,
|
| 692 |
+
"loss": 0.1854,
|
| 693 |
+
"step": 1775
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 0.65,
|
| 697 |
+
"grad_norm": 4.869482040405273,
|
| 698 |
+
"learning_rate": 1.1355769230769231e-05,
|
| 699 |
+
"loss": 0.1962,
|
| 700 |
+
"step": 1800
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 0.65,
|
| 704 |
+
"eval_loss": 0.22110989689826965,
|
| 705 |
+
"eval_na_accuracy": 0.9140893220901489,
|
| 706 |
+
"eval_ordinal_accuracy": 0.6231697797775269,
|
| 707 |
+
"eval_ordinal_mae": 0.4857858419418335,
|
| 708 |
+
"eval_runtime": 187.6432,
|
| 709 |
+
"eval_samples_per_second": 23.848,
|
| 710 |
+
"eval_steps_per_second": 2.984,
|
| 711 |
+
"step": 1800
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.66,
|
| 715 |
+
"grad_norm": 5.728977203369141,
|
| 716 |
+
"learning_rate": 1.1235576923076923e-05,
|
| 717 |
+
"loss": 0.2294,
|
| 718 |
+
"step": 1825
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.67,
|
| 722 |
+
"grad_norm": 0.5772213339805603,
|
| 723 |
+
"learning_rate": 1.1115384615384617e-05,
|
| 724 |
+
"loss": 0.1981,
|
| 725 |
+
"step": 1850
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.68,
|
| 729 |
+
"grad_norm": 2.899949550628662,
|
| 730 |
+
"learning_rate": 1.0995192307692308e-05,
|
| 731 |
+
"loss": 0.1885,
|
| 732 |
+
"step": 1875
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.69,
|
| 736 |
+
"grad_norm": 1.060448408126831,
|
| 737 |
+
"learning_rate": 1.0875e-05,
|
| 738 |
+
"loss": 0.1882,
|
| 739 |
+
"step": 1900
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.69,
|
| 743 |
+
"eval_loss": 0.20448338985443115,
|
| 744 |
+
"eval_na_accuracy": 0.8625429272651672,
|
| 745 |
+
"eval_ordinal_accuracy": 0.6632417440414429,
|
| 746 |
+
"eval_ordinal_mae": 0.4668627381324768,
|
| 747 |
+
"eval_runtime": 187.7227,
|
| 748 |
+
"eval_samples_per_second": 23.838,
|
| 749 |
+
"eval_steps_per_second": 2.983,
|
| 750 |
+
"step": 1900
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.69,
|
| 754 |
+
"grad_norm": 8.596738815307617,
|
| 755 |
+
"learning_rate": 1.0754807692307693e-05,
|
| 756 |
+
"loss": 0.2388,
|
| 757 |
+
"step": 1925
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.7,
|
| 761 |
+
"grad_norm": 2.922056198120117,
|
| 762 |
+
"learning_rate": 1.0634615384615385e-05,
|
| 763 |
+
"loss": 0.2172,
|
| 764 |
+
"step": 1950
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.71,
|
| 768 |
+
"grad_norm": 2.6090290546417236,
|
| 769 |
+
"learning_rate": 1.0514423076923077e-05,
|
| 770 |
+
"loss": 0.2432,
|
| 771 |
+
"step": 1975
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.72,
|
| 775 |
+
"grad_norm": 0.9129126667976379,
|
| 776 |
+
"learning_rate": 1.039423076923077e-05,
|
| 777 |
+
"loss": 0.1895,
|
| 778 |
+
"step": 2000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.72,
|
| 782 |
+
"eval_loss": 0.2081986963748932,
|
| 783 |
+
"eval_na_accuracy": 0.8608247637748718,
|
| 784 |
+
"eval_ordinal_accuracy": 0.6316465735435486,
|
| 785 |
+
"eval_ordinal_mae": 0.46963009238243103,
|
| 786 |
+
"eval_runtime": 177.2649,
|
| 787 |
+
"eval_samples_per_second": 25.245,
|
| 788 |
+
"eval_steps_per_second": 3.159,
|
| 789 |
+
"step": 2000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.73,
|
| 793 |
+
"grad_norm": 14.650406837463379,
|
| 794 |
+
"learning_rate": 1.0274038461538462e-05,
|
| 795 |
+
"loss": 0.2168,
|
| 796 |
+
"step": 2025
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.74,
|
| 800 |
+
"grad_norm": 2.405910015106201,
|
| 801 |
+
"learning_rate": 1.0153846153846154e-05,
|
| 802 |
+
"loss": 0.224,
|
| 803 |
+
"step": 2050
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.75,
|
| 807 |
+
"grad_norm": 1.4129964113235474,
|
| 808 |
+
"learning_rate": 1.0033653846153847e-05,
|
| 809 |
+
"loss": 0.1908,
|
| 810 |
+
"step": 2075
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.76,
|
| 814 |
+
"grad_norm": 2.485114812850952,
|
| 815 |
+
"learning_rate": 9.913461538461539e-06,
|
| 816 |
+
"loss": 0.1979,
|
| 817 |
+
"step": 2100
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.76,
|
| 821 |
+
"eval_loss": 0.22696280479431152,
|
| 822 |
+
"eval_na_accuracy": 0.900343656539917,
|
| 823 |
+
"eval_ordinal_accuracy": 0.6372976899147034,
|
| 824 |
+
"eval_ordinal_mae": 0.4791434407234192,
|
| 825 |
+
"eval_runtime": 187.006,
|
| 826 |
+
"eval_samples_per_second": 23.93,
|
| 827 |
+
"eval_steps_per_second": 2.995,
|
| 828 |
+
"step": 2100
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.77,
|
| 832 |
+
"grad_norm": 1.7261921167373657,
|
| 833 |
+
"learning_rate": 9.79326923076923e-06,
|
| 834 |
+
"loss": 0.2169,
|
| 835 |
+
"step": 2125
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.78,
|
| 839 |
+
"grad_norm": 5.685389995574951,
|
| 840 |
+
"learning_rate": 9.673076923076924e-06,
|
| 841 |
+
"loss": 0.2446,
|
| 842 |
+
"step": 2150
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.78,
|
| 846 |
+
"grad_norm": 1.2241212129592896,
|
| 847 |
+
"learning_rate": 9.552884615384616e-06,
|
| 848 |
+
"loss": 0.2202,
|
| 849 |
+
"step": 2175
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.79,
|
| 853 |
+
"grad_norm": 3.1511054039001465,
|
| 854 |
+
"learning_rate": 9.432692307692308e-06,
|
| 855 |
+
"loss": 0.2643,
|
| 856 |
+
"step": 2200
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.79,
|
| 860 |
+
"eval_loss": 0.20690996944904327,
|
| 861 |
+
"eval_na_accuracy": 0.8556700944900513,
|
| 862 |
+
"eval_ordinal_accuracy": 0.6414076685905457,
|
| 863 |
+
"eval_ordinal_mae": 0.46626007556915283,
|
| 864 |
+
"eval_runtime": 194.378,
|
| 865 |
+
"eval_samples_per_second": 23.022,
|
| 866 |
+
"eval_steps_per_second": 2.881,
|
| 867 |
+
"step": 2200
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.8,
|
| 871 |
+
"grad_norm": 2.382234811782837,
|
| 872 |
+
"learning_rate": 9.312500000000001e-06,
|
| 873 |
+
"loss": 0.1782,
|
| 874 |
+
"step": 2225
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.81,
|
| 878 |
+
"grad_norm": 3.646544933319092,
|
| 879 |
+
"learning_rate": 9.192307692307693e-06,
|
| 880 |
+
"loss": 0.1901,
|
| 881 |
+
"step": 2250
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.82,
|
| 885 |
+
"grad_norm": 3.5835981369018555,
|
| 886 |
+
"learning_rate": 9.072115384615385e-06,
|
| 887 |
+
"loss": 0.179,
|
| 888 |
+
"step": 2275
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.83,
|
| 892 |
+
"grad_norm": 0.6391886472702026,
|
| 893 |
+
"learning_rate": 8.951923076923078e-06,
|
| 894 |
+
"loss": 0.2279,
|
| 895 |
+
"step": 2300
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.83,
|
| 899 |
+
"eval_loss": 0.2029835283756256,
|
| 900 |
+
"eval_na_accuracy": 0.869415819644928,
|
| 901 |
+
"eval_ordinal_accuracy": 0.654251217842102,
|
| 902 |
+
"eval_ordinal_mae": 0.4581436216831207,
|
| 903 |
+
"eval_runtime": 190.8034,
|
| 904 |
+
"eval_samples_per_second": 23.453,
|
| 905 |
+
"eval_steps_per_second": 2.935,
|
| 906 |
+
"step": 2300
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.84,
|
| 910 |
+
"grad_norm": 1.8583753108978271,
|
| 911 |
+
"learning_rate": 8.83173076923077e-06,
|
| 912 |
+
"loss": 0.2403,
|
| 913 |
+
"step": 2325
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.85,
|
| 917 |
+
"grad_norm": 7.337312698364258,
|
| 918 |
+
"learning_rate": 8.711538461538463e-06,
|
| 919 |
+
"loss": 0.2262,
|
| 920 |
+
"step": 2350
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.86,
|
| 924 |
+
"grad_norm": 4.292835712432861,
|
| 925 |
+
"learning_rate": 8.591346153846155e-06,
|
| 926 |
+
"loss": 0.2321,
|
| 927 |
+
"step": 2375
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.87,
|
| 931 |
+
"grad_norm": 3.3267788887023926,
|
| 932 |
+
"learning_rate": 8.471153846153847e-06,
|
| 933 |
+
"loss": 0.1965,
|
| 934 |
+
"step": 2400
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.87,
|
| 938 |
+
"eval_loss": 0.21094879508018494,
|
| 939 |
+
"eval_na_accuracy": 0.800687313079834,
|
| 940 |
+
"eval_ordinal_accuracy": 0.681993305683136,
|
| 941 |
+
"eval_ordinal_mae": 0.44459667801856995,
|
| 942 |
+
"eval_runtime": 184.1529,
|
| 943 |
+
"eval_samples_per_second": 24.3,
|
| 944 |
+
"eval_steps_per_second": 3.041,
|
| 945 |
+
"step": 2400
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.87,
|
| 949 |
+
"grad_norm": 2.07079815864563,
|
| 950 |
+
"learning_rate": 8.35096153846154e-06,
|
| 951 |
+
"loss": 0.2122,
|
| 952 |
+
"step": 2425
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.88,
|
| 956 |
+
"grad_norm": 3.6883444786071777,
|
| 957 |
+
"learning_rate": 8.230769230769232e-06,
|
| 958 |
+
"loss": 0.1876,
|
| 959 |
+
"step": 2450
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.89,
|
| 963 |
+
"grad_norm": 0.5956806540489197,
|
| 964 |
+
"learning_rate": 8.110576923076923e-06,
|
| 965 |
+
"loss": 0.1804,
|
| 966 |
+
"step": 2475
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 0.9,
|
| 970 |
+
"grad_norm": 0.6182098984718323,
|
| 971 |
+
"learning_rate": 7.990384615384617e-06,
|
| 972 |
+
"loss": 0.1637,
|
| 973 |
+
"step": 2500
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 0.9,
|
| 977 |
+
"eval_loss": 0.20052286982536316,
|
| 978 |
+
"eval_na_accuracy": 0.8556700944900513,
|
| 979 |
+
"eval_ordinal_accuracy": 0.6763421297073364,
|
| 980 |
+
"eval_ordinal_mae": 0.4438597857952118,
|
| 981 |
+
"eval_runtime": 183.167,
|
| 982 |
+
"eval_samples_per_second": 24.431,
|
| 983 |
+
"eval_steps_per_second": 3.057,
|
| 984 |
+
"step": 2500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.91,
|
| 988 |
+
"grad_norm": 2.0849900245666504,
|
| 989 |
+
"learning_rate": 7.875e-06,
|
| 990 |
+
"loss": 0.2287,
|
| 991 |
+
"step": 2525
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.92,
|
| 995 |
+
"grad_norm": 2.9747681617736816,
|
| 996 |
+
"learning_rate": 7.754807692307693e-06,
|
| 997 |
+
"loss": 0.202,
|
| 998 |
+
"step": 2550
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.93,
|
| 1002 |
+
"grad_norm": 0.7342644333839417,
|
| 1003 |
+
"learning_rate": 7.634615384615385e-06,
|
| 1004 |
+
"loss": 0.1887,
|
| 1005 |
+
"step": 2575
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.94,
|
| 1009 |
+
"grad_norm": 2.5988609790802,
|
| 1010 |
+
"learning_rate": 7.514423076923078e-06,
|
| 1011 |
+
"loss": 0.1705,
|
| 1012 |
+
"step": 2600
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.94,
|
| 1016 |
+
"eval_loss": 0.19641266763210297,
|
| 1017 |
+
"eval_na_accuracy": 0.8539518713951111,
|
| 1018 |
+
"eval_ordinal_accuracy": 0.6748009324073792,
|
| 1019 |
+
"eval_ordinal_mae": 0.43212634325027466,
|
| 1020 |
+
"eval_runtime": 189.9978,
|
| 1021 |
+
"eval_samples_per_second": 23.553,
|
| 1022 |
+
"eval_steps_per_second": 2.947,
|
| 1023 |
+
"step": 2600
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.95,
|
| 1027 |
+
"grad_norm": 3.347687244415283,
|
| 1028 |
+
"learning_rate": 7.39423076923077e-06,
|
| 1029 |
+
"loss": 0.1947,
|
| 1030 |
+
"step": 2625
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.96,
|
| 1034 |
+
"grad_norm": 1.516992449760437,
|
| 1035 |
+
"learning_rate": 7.274038461538462e-06,
|
| 1036 |
+
"loss": 0.2137,
|
| 1037 |
+
"step": 2650
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.96,
|
| 1041 |
+
"grad_norm": 1.79114830493927,
|
| 1042 |
+
"learning_rate": 7.153846153846155e-06,
|
| 1043 |
+
"loss": 0.2398,
|
| 1044 |
+
"step": 2675
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.97,
|
| 1048 |
+
"grad_norm": 2.8799991607666016,
|
| 1049 |
+
"learning_rate": 7.033653846153847e-06,
|
| 1050 |
+
"loss": 0.2412,
|
| 1051 |
+
"step": 2700
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.97,
|
| 1055 |
+
"eval_loss": 0.19578155875205994,
|
| 1056 |
+
"eval_na_accuracy": 0.8780068755149841,
|
| 1057 |
+
"eval_ordinal_accuracy": 0.6730028390884399,
|
| 1058 |
+
"eval_ordinal_mae": 0.4344838559627533,
|
| 1059 |
+
"eval_runtime": 198.8509,
|
| 1060 |
+
"eval_samples_per_second": 22.504,
|
| 1061 |
+
"eval_steps_per_second": 2.816,
|
| 1062 |
+
"step": 2700
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 0.98,
|
| 1066 |
+
"grad_norm": 1.0458216667175293,
|
| 1067 |
+
"learning_rate": 6.913461538461539e-06,
|
| 1068 |
+
"loss": 0.1697,
|
| 1069 |
+
"step": 2725
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 0.99,
|
| 1073 |
+
"grad_norm": 2.2299771308898926,
|
| 1074 |
+
"learning_rate": 6.7932692307692315e-06,
|
| 1075 |
+
"loss": 0.1623,
|
| 1076 |
+
"step": 2750
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.0,
|
| 1080 |
+
"grad_norm": 1.825832724571228,
|
| 1081 |
+
"learning_rate": 6.673076923076923e-06,
|
| 1082 |
+
"loss": 0.2052,
|
| 1083 |
+
"step": 2775
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 1.01,
|
| 1087 |
+
"grad_norm": 0.6158725619316101,
|
| 1088 |
+
"learning_rate": 6.552884615384616e-06,
|
| 1089 |
+
"loss": 0.1438,
|
| 1090 |
+
"step": 2800
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.01,
|
| 1094 |
+
"eval_loss": 0.19719891250133514,
|
| 1095 |
+
"eval_na_accuracy": 0.8470790386199951,
|
| 1096 |
+
"eval_ordinal_accuracy": 0.6783971190452576,
|
| 1097 |
+
"eval_ordinal_mae": 0.43012040853500366,
|
| 1098 |
+
"eval_runtime": 190.7061,
|
| 1099 |
+
"eval_samples_per_second": 23.465,
|
| 1100 |
+
"eval_steps_per_second": 2.936,
|
| 1101 |
+
"step": 2800
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.02,
|
| 1105 |
+
"grad_norm": 1.041905164718628,
|
| 1106 |
+
"learning_rate": 6.432692307692308e-06,
|
| 1107 |
+
"loss": 0.1584,
|
| 1108 |
+
"step": 2825
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.03,
|
| 1112 |
+
"grad_norm": 0.6891164779663086,
|
| 1113 |
+
"learning_rate": 6.3125e-06,
|
| 1114 |
+
"loss": 0.1488,
|
| 1115 |
+
"step": 2850
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.04,
|
| 1119 |
+
"grad_norm": 4.814165115356445,
|
| 1120 |
+
"learning_rate": 6.192307692307693e-06,
|
| 1121 |
+
"loss": 0.1565,
|
| 1122 |
+
"step": 2875
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.05,
|
| 1126 |
+
"grad_norm": 1.0739597082138062,
|
| 1127 |
+
"learning_rate": 6.0721153846153844e-06,
|
| 1128 |
+
"loss": 0.123,
|
| 1129 |
+
"step": 2900
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.05,
|
| 1133 |
+
"eval_loss": 0.19954617321491241,
|
| 1134 |
+
"eval_na_accuracy": 0.8419243693351746,
|
| 1135 |
+
"eval_ordinal_accuracy": 0.6753146648406982,
|
| 1136 |
+
"eval_ordinal_mae": 0.42309799790382385,
|
| 1137 |
+
"eval_runtime": 185.744,
|
| 1138 |
+
"eval_samples_per_second": 24.092,
|
| 1139 |
+
"eval_steps_per_second": 3.015,
|
| 1140 |
+
"step": 2900
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.05,
|
| 1144 |
+
"grad_norm": 3.492755651473999,
|
| 1145 |
+
"learning_rate": 5.951923076923077e-06,
|
| 1146 |
+
"loss": 0.1547,
|
| 1147 |
+
"step": 2925
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 1.06,
|
| 1151 |
+
"grad_norm": 0.9093284010887146,
|
| 1152 |
+
"learning_rate": 5.8317307692307704e-06,
|
| 1153 |
+
"loss": 0.1258,
|
| 1154 |
+
"step": 2950
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.07,
|
| 1158 |
+
"grad_norm": 0.7456061244010925,
|
| 1159 |
+
"learning_rate": 5.711538461538461e-06,
|
| 1160 |
+
"loss": 0.1648,
|
| 1161 |
+
"step": 2975
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 1.08,
|
| 1165 |
+
"grad_norm": 0.8997055888175964,
|
| 1166 |
+
"learning_rate": 5.591346153846155e-06,
|
| 1167 |
+
"loss": 0.1411,
|
| 1168 |
+
"step": 3000
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.08,
|
| 1172 |
+
"eval_loss": 0.19463004171848297,
|
| 1173 |
+
"eval_na_accuracy": 0.8453608155250549,
|
| 1174 |
+
"eval_ordinal_accuracy": 0.6817364692687988,
|
| 1175 |
+
"eval_ordinal_mae": 0.42202073335647583,
|
| 1176 |
+
"eval_runtime": 176.571,
|
| 1177 |
+
"eval_samples_per_second": 25.344,
|
| 1178 |
+
"eval_steps_per_second": 3.172,
|
| 1179 |
+
"step": 3000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.09,
|
| 1183 |
+
"grad_norm": 0.5158917903900146,
|
| 1184 |
+
"learning_rate": 5.471153846153847e-06,
|
| 1185 |
+
"loss": 0.168,
|
| 1186 |
+
"step": 3025
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.1,
|
| 1190 |
+
"grad_norm": 1.5927814245224,
|
| 1191 |
+
"learning_rate": 5.350961538461539e-06,
|
| 1192 |
+
"loss": 0.1113,
|
| 1193 |
+
"step": 3050
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.11,
|
| 1197 |
+
"grad_norm": 0.7580274343490601,
|
| 1198 |
+
"learning_rate": 5.230769230769232e-06,
|
| 1199 |
+
"loss": 0.1489,
|
| 1200 |
+
"step": 3075
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.12,
|
| 1204 |
+
"grad_norm": 0.6561002135276794,
|
| 1205 |
+
"learning_rate": 5.110576923076923e-06,
|
| 1206 |
+
"loss": 0.1443,
|
| 1207 |
+
"step": 3100
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.12,
|
| 1211 |
+
"eval_loss": 0.19156721234321594,
|
| 1212 |
+
"eval_na_accuracy": 0.8591065406799316,
|
| 1213 |
+
"eval_ordinal_accuracy": 0.682763934135437,
|
| 1214 |
+
"eval_ordinal_mae": 0.42213648557662964,
|
| 1215 |
+
"eval_runtime": 180.8137,
|
| 1216 |
+
"eval_samples_per_second": 24.749,
|
| 1217 |
+
"eval_steps_per_second": 3.097,
|
| 1218 |
+
"step": 3100
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 1.13,
|
| 1222 |
+
"grad_norm": 0.7529481053352356,
|
| 1223 |
+
"learning_rate": 4.990384615384616e-06,
|
| 1224 |
+
"loss": 0.1546,
|
| 1225 |
+
"step": 3125
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 1.14,
|
| 1229 |
+
"grad_norm": 3.369683027267456,
|
| 1230 |
+
"learning_rate": 4.870192307692308e-06,
|
| 1231 |
+
"loss": 0.136,
|
| 1232 |
+
"step": 3150
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.14,
|
| 1236 |
+
"grad_norm": 0.9865265488624573,
|
| 1237 |
+
"learning_rate": 4.75e-06,
|
| 1238 |
+
"loss": 0.1498,
|
| 1239 |
+
"step": 3175
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 1.15,
|
| 1243 |
+
"grad_norm": 1.8516024351119995,
|
| 1244 |
+
"learning_rate": 4.629807692307693e-06,
|
| 1245 |
+
"loss": 0.208,
|
| 1246 |
+
"step": 3200
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.15,
|
| 1250 |
+
"eval_loss": 0.19419582188129425,
|
| 1251 |
+
"eval_na_accuracy": 0.8676975965499878,
|
| 1252 |
+
"eval_ordinal_accuracy": 0.6740303039550781,
|
| 1253 |
+
"eval_ordinal_mae": 0.4163132607936859,
|
| 1254 |
+
"eval_runtime": 188.6406,
|
| 1255 |
+
"eval_samples_per_second": 23.722,
|
| 1256 |
+
"eval_steps_per_second": 2.969,
|
| 1257 |
+
"step": 3200
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 1.16,
|
| 1261 |
+
"grad_norm": 1.1575658321380615,
|
| 1262 |
+
"learning_rate": 4.509615384615385e-06,
|
| 1263 |
+
"loss": 0.1316,
|
| 1264 |
+
"step": 3225
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.17,
|
| 1268 |
+
"grad_norm": 0.9484291076660156,
|
| 1269 |
+
"learning_rate": 4.389423076923077e-06,
|
| 1270 |
+
"loss": 0.1204,
|
| 1271 |
+
"step": 3250
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 1.18,
|
| 1275 |
+
"grad_norm": 2.565762519836426,
|
| 1276 |
+
"learning_rate": 4.26923076923077e-06,
|
| 1277 |
+
"loss": 0.1262,
|
| 1278 |
+
"step": 3275
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 1.19,
|
| 1282 |
+
"grad_norm": 2.2757420539855957,
|
| 1283 |
+
"learning_rate": 4.149038461538462e-06,
|
| 1284 |
+
"loss": 0.1343,
|
| 1285 |
+
"step": 3300
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.19,
|
| 1289 |
+
"eval_loss": 0.19619733095169067,
|
| 1290 |
+
"eval_na_accuracy": 0.8470790386199951,
|
| 1291 |
+
"eval_ordinal_accuracy": 0.6889288425445557,
|
| 1292 |
+
"eval_ordinal_mae": 0.4182307720184326,
|
| 1293 |
+
"eval_runtime": 187.8854,
|
| 1294 |
+
"eval_samples_per_second": 23.818,
|
| 1295 |
+
"eval_steps_per_second": 2.981,
|
| 1296 |
+
"step": 3300
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.2,
|
| 1300 |
+
"grad_norm": 3.402385711669922,
|
| 1301 |
+
"learning_rate": 4.028846153846154e-06,
|
| 1302 |
+
"loss": 0.1397,
|
| 1303 |
+
"step": 3325
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 1.21,
|
| 1307 |
+
"grad_norm": 0.881535530090332,
|
| 1308 |
+
"learning_rate": 3.908653846153847e-06,
|
| 1309 |
+
"loss": 0.1176,
|
| 1310 |
+
"step": 3350
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 1.22,
|
| 1314 |
+
"grad_norm": 0.948428213596344,
|
| 1315 |
+
"learning_rate": 3.7884615384615388e-06,
|
| 1316 |
+
"loss": 0.1767,
|
| 1317 |
+
"step": 3375
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 1.23,
|
| 1321 |
+
"grad_norm": 1.2142385244369507,
|
| 1322 |
+
"learning_rate": 3.668269230769231e-06,
|
| 1323 |
+
"loss": 0.1347,
|
| 1324 |
+
"step": 3400
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 1.23,
|
| 1328 |
+
"eval_loss": 0.1938188225030899,
|
| 1329 |
+
"eval_na_accuracy": 0.8659793734550476,
|
| 1330 |
+
"eval_ordinal_accuracy": 0.6899563074111938,
|
| 1331 |
+
"eval_ordinal_mae": 0.4161255955696106,
|
| 1332 |
+
"eval_runtime": 185.9367,
|
| 1333 |
+
"eval_samples_per_second": 24.067,
|
| 1334 |
+
"eval_steps_per_second": 3.012,
|
| 1335 |
+
"step": 3400
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 1.24,
|
| 1339 |
+
"grad_norm": 3.2418200969696045,
|
| 1340 |
+
"learning_rate": 3.5480769230769235e-06,
|
| 1341 |
+
"loss": 0.1585,
|
| 1342 |
+
"step": 3425
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.24,
|
| 1346 |
+
"grad_norm": 1.845564842224121,
|
| 1347 |
+
"learning_rate": 3.4278846153846157e-06,
|
| 1348 |
+
"loss": 0.1499,
|
| 1349 |
+
"step": 3450
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.25,
|
| 1353 |
+
"grad_norm": 1.7326797246932983,
|
| 1354 |
+
"learning_rate": 3.307692307692308e-06,
|
| 1355 |
+
"loss": 0.1286,
|
| 1356 |
+
"step": 3475
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.26,
|
| 1360 |
+
"grad_norm": 4.901269912719727,
|
| 1361 |
+
"learning_rate": 3.1875e-06,
|
| 1362 |
+
"loss": 0.1076,
|
| 1363 |
+
"step": 3500
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.26,
|
| 1367 |
+
"eval_loss": 0.19698283076286316,
|
| 1368 |
+
"eval_na_accuracy": 0.8470790386199951,
|
| 1369 |
+
"eval_ordinal_accuracy": 0.6943231225013733,
|
| 1370 |
+
"eval_ordinal_mae": 0.41806870698928833,
|
| 1371 |
+
"eval_runtime": 187.7833,
|
| 1372 |
+
"eval_samples_per_second": 23.831,
|
| 1373 |
+
"eval_steps_per_second": 2.982,
|
| 1374 |
+
"step": 3500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.27,
|
| 1378 |
+
"grad_norm": 1.3833277225494385,
|
| 1379 |
+
"learning_rate": 3.0673076923076926e-06,
|
| 1380 |
+
"loss": 0.1246,
|
| 1381 |
+
"step": 3525
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.28,
|
| 1385 |
+
"grad_norm": 0.9357690215110779,
|
| 1386 |
+
"learning_rate": 2.947115384615385e-06,
|
| 1387 |
+
"loss": 0.1113,
|
| 1388 |
+
"step": 3550
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.29,
|
| 1392 |
+
"grad_norm": 7.975602149963379,
|
| 1393 |
+
"learning_rate": 2.8269230769230773e-06,
|
| 1394 |
+
"loss": 0.1872,
|
| 1395 |
+
"step": 3575
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.3,
|
| 1399 |
+
"grad_norm": 0.5384923815727234,
|
| 1400 |
+
"learning_rate": 2.7067307692307694e-06,
|
| 1401 |
+
"loss": 0.1248,
|
| 1402 |
+
"step": 3600
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.3,
|
| 1406 |
+
"eval_loss": 0.19512778520584106,
|
| 1407 |
+
"eval_na_accuracy": 0.8470790386199951,
|
| 1408 |
+
"eval_ordinal_accuracy": 0.6958643794059753,
|
| 1409 |
+
"eval_ordinal_mae": 0.4150661826133728,
|
| 1410 |
+
"eval_runtime": 190.3088,
|
| 1411 |
+
"eval_samples_per_second": 23.514,
|
| 1412 |
+
"eval_steps_per_second": 2.943,
|
| 1413 |
+
"step": 3600
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 1.31,
|
| 1417 |
+
"grad_norm": 0.7750712633132935,
|
| 1418 |
+
"learning_rate": 2.586538461538462e-06,
|
| 1419 |
+
"loss": 0.1249,
|
| 1420 |
+
"step": 3625
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 1.32,
|
| 1424 |
+
"grad_norm": 0.7317385077476501,
|
| 1425 |
+
"learning_rate": 2.466346153846154e-06,
|
| 1426 |
+
"loss": 0.124,
|
| 1427 |
+
"step": 3650
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 1.33,
|
| 1431 |
+
"grad_norm": 0.79685378074646,
|
| 1432 |
+
"learning_rate": 2.3461538461538463e-06,
|
| 1433 |
+
"loss": 0.1524,
|
| 1434 |
+
"step": 3675
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 1.33,
|
| 1438 |
+
"grad_norm": 0.9845700263977051,
|
| 1439 |
+
"learning_rate": 2.2259615384615385e-06,
|
| 1440 |
+
"loss": 0.1455,
|
| 1441 |
+
"step": 3700
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.33,
|
| 1445 |
+
"eval_loss": 0.1951962560415268,
|
| 1446 |
+
"eval_na_accuracy": 0.8814433217048645,
|
| 1447 |
+
"eval_ordinal_accuracy": 0.6850757598876953,
|
| 1448 |
+
"eval_ordinal_mae": 0.4146950840950012,
|
| 1449 |
+
"eval_runtime": 188.1776,
|
| 1450 |
+
"eval_samples_per_second": 23.781,
|
| 1451 |
+
"eval_steps_per_second": 2.976,
|
| 1452 |
+
"step": 3700
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.34,
|
| 1456 |
+
"grad_norm": 2.780651330947876,
|
| 1457 |
+
"learning_rate": 2.105769230769231e-06,
|
| 1458 |
+
"loss": 0.174,
|
| 1459 |
+
"step": 3725
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.35,
|
| 1463 |
+
"grad_norm": 1.0597649812698364,
|
| 1464 |
+
"learning_rate": 1.9855769230769232e-06,
|
| 1465 |
+
"loss": 0.1211,
|
| 1466 |
+
"step": 3750
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.36,
|
| 1470 |
+
"grad_norm": 2.3485703468322754,
|
| 1471 |
+
"learning_rate": 1.8653846153846156e-06,
|
| 1472 |
+
"loss": 0.1537,
|
| 1473 |
+
"step": 3775
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.37,
|
| 1477 |
+
"grad_norm": 6.283831596374512,
|
| 1478 |
+
"learning_rate": 1.7451923076923077e-06,
|
| 1479 |
+
"loss": 0.131,
|
| 1480 |
+
"step": 3800
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.37,
|
| 1484 |
+
"eval_loss": 0.19528667628765106,
|
| 1485 |
+
"eval_na_accuracy": 0.8453608155250549,
|
| 1486 |
+
"eval_ordinal_accuracy": 0.6948369145393372,
|
| 1487 |
+
"eval_ordinal_mae": 0.4172358810901642,
|
| 1488 |
+
"eval_runtime": 191.9622,
|
| 1489 |
+
"eval_samples_per_second": 23.312,
|
| 1490 |
+
"eval_steps_per_second": 2.917,
|
| 1491 |
+
"step": 3800
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 1.38,
|
| 1495 |
+
"grad_norm": 0.6940748691558838,
|
| 1496 |
+
"learning_rate": 1.6250000000000001e-06,
|
| 1497 |
+
"loss": 0.1321,
|
| 1498 |
+
"step": 3825
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.39,
|
| 1502 |
+
"grad_norm": 2.970365285873413,
|
| 1503 |
+
"learning_rate": 1.5048076923076923e-06,
|
| 1504 |
+
"loss": 0.1578,
|
| 1505 |
+
"step": 3850
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.4,
|
| 1509 |
+
"grad_norm": 3.247617483139038,
|
| 1510 |
+
"learning_rate": 1.3846153846153848e-06,
|
| 1511 |
+
"loss": 0.1484,
|
| 1512 |
+
"step": 3875
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.41,
|
| 1516 |
+
"grad_norm": 0.7711169719696045,
|
| 1517 |
+
"learning_rate": 1.264423076923077e-06,
|
| 1518 |
+
"loss": 0.1307,
|
| 1519 |
+
"step": 3900
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 1.41,
|
| 1523 |
+
"eval_loss": 0.19318200647830963,
|
| 1524 |
+
"eval_na_accuracy": 0.8642611503601074,
|
| 1525 |
+
"eval_ordinal_accuracy": 0.692781925201416,
|
| 1526 |
+
"eval_ordinal_mae": 0.41271111369132996,
|
| 1527 |
+
"eval_runtime": 192.6809,
|
| 1528 |
+
"eval_samples_per_second": 23.225,
|
| 1529 |
+
"eval_steps_per_second": 2.906,
|
| 1530 |
+
"step": 3900
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 1.42,
|
| 1534 |
+
"grad_norm": 0.6948511004447937,
|
| 1535 |
+
"learning_rate": 1.1442307692307694e-06,
|
| 1536 |
+
"loss": 0.1566,
|
| 1537 |
+
"step": 3925
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.42,
|
| 1541 |
+
"grad_norm": 0.9776670336723328,
|
| 1542 |
+
"learning_rate": 1.0240384615384615e-06,
|
| 1543 |
+
"loss": 0.153,
|
| 1544 |
+
"step": 3950
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.43,
|
| 1548 |
+
"grad_norm": 1.5596826076507568,
|
| 1549 |
+
"learning_rate": 9.038461538461539e-07,
|
| 1550 |
+
"loss": 0.1316,
|
| 1551 |
+
"step": 3975
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 1.44,
|
| 1555 |
+
"grad_norm": 0.5882940888404846,
|
| 1556 |
+
"learning_rate": 7.836538461538463e-07,
|
| 1557 |
+
"loss": 0.1198,
|
| 1558 |
+
"step": 4000
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 1.44,
|
| 1562 |
+
"eval_loss": 0.19471855461597443,
|
| 1563 |
+
"eval_na_accuracy": 0.8573883175849915,
|
| 1564 |
+
"eval_ordinal_accuracy": 0.6940662860870361,
|
| 1565 |
+
"eval_ordinal_mae": 0.4110487997531891,
|
| 1566 |
+
"eval_runtime": 192.5912,
|
| 1567 |
+
"eval_samples_per_second": 23.236,
|
| 1568 |
+
"eval_steps_per_second": 2.908,
|
| 1569 |
+
"step": 4000
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 1.45,
|
| 1573 |
+
"grad_norm": 1.4058364629745483,
|
| 1574 |
+
"learning_rate": 6.634615384615385e-07,
|
| 1575 |
+
"loss": 0.1154,
|
| 1576 |
+
"step": 4025
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.46,
|
| 1580 |
+
"grad_norm": 2.0241715908050537,
|
| 1581 |
+
"learning_rate": 5.432692307692308e-07,
|
| 1582 |
+
"loss": 0.1359,
|
| 1583 |
+
"step": 4050
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 1.47,
|
| 1587 |
+
"grad_norm": 2.556328773498535,
|
| 1588 |
+
"learning_rate": 4.2307692307692315e-07,
|
| 1589 |
+
"loss": 0.1374,
|
| 1590 |
+
"step": 4075
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 1.48,
|
| 1594 |
+
"grad_norm": 0.89422607421875,
|
| 1595 |
+
"learning_rate": 3.028846153846154e-07,
|
| 1596 |
+
"loss": 0.1363,
|
| 1597 |
+
"step": 4100
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 1.48,
|
| 1601 |
+
"eval_loss": 0.19520333409309387,
|
| 1602 |
+
"eval_na_accuracy": 0.8573883175849915,
|
| 1603 |
+
"eval_ordinal_accuracy": 0.6886719465255737,
|
| 1604 |
+
"eval_ordinal_mae": 0.4086832106113434,
|
| 1605 |
+
"eval_runtime": 191.6181,
|
| 1606 |
+
"eval_samples_per_second": 23.354,
|
| 1607 |
+
"eval_steps_per_second": 2.922,
|
| 1608 |
+
"step": 4100
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.49,
|
| 1612 |
+
"grad_norm": 1.4230666160583496,
|
| 1613 |
+
"learning_rate": 1.8269230769230772e-07,
|
| 1614 |
+
"loss": 0.1273,
|
| 1615 |
+
"step": 4125
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 1.5,
|
| 1619 |
+
"grad_norm": 2.7363412380218506,
|
| 1620 |
+
"learning_rate": 6.250000000000001e-08,
|
| 1621 |
+
"loss": 0.1068,
|
| 1622 |
+
"step": 4150
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 1.5,
|
| 1626 |
+
"step": 4160,
|
| 1627 |
+
"total_flos": 5.158051742063002e+18,
|
| 1628 |
+
"train_loss": 0.21532289660893955,
|
| 1629 |
+
"train_runtime": 13643.9574,
|
| 1630 |
+
"train_samples_per_second": 4.878,
|
| 1631 |
+
"train_steps_per_second": 0.305
|
| 1632 |
+
}
|
| 1633 |
+
],
|
| 1634 |
+
"logging_steps": 25,
|
| 1635 |
+
"max_steps": 4160,
|
| 1636 |
+
"num_input_tokens_seen": 0,
|
| 1637 |
+
"num_train_epochs": 2,
|
| 1638 |
+
"save_steps": 100,
|
| 1639 |
+
"total_flos": 5.158051742063002e+18,
|
| 1640 |
+
"train_batch_size": 16,
|
| 1641 |
+
"trial_name": null,
|
| 1642 |
+
"trial_params": null
|
| 1643 |
+
}
|