🍻 cheers
Browse files- README.md +6 -5
- all_results.json +14 -0
- eval_results.json +10 -0
- runs/Mar31_17-12-55_ryanserver/events.out.tfevents.1711937747.ryanserver.31714.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +2182 -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_v2
|
|
@@ -13,12 +14,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# ryan03312024_lr_2e-5_wd_001_v2
|
| 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_v2
|
|
|
|
| 14 |
|
| 15 |
# ryan03312024_lr_2e-5_wd_001_v2
|
| 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: 0.1914
|
| 20 |
+
- Ordinal Mae: 0.4198
|
| 21 |
+
- Ordinal Accuracy: 0.6843
|
| 22 |
+
- Na Accuracy: 0.8505
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_loss": 0.1913878321647644,
|
| 4 |
+
"eval_na_accuracy": 0.8505154848098755,
|
| 5 |
+
"eval_ordinal_accuracy": 0.6843051910400391,
|
| 6 |
+
"eval_ordinal_mae": 0.41984713077545166,
|
| 7 |
+
"eval_runtime": 173.1798,
|
| 8 |
+
"eval_samples_per_second": 25.84,
|
| 9 |
+
"eval_steps_per_second": 3.234,
|
| 10 |
+
"train_loss": 0.18857605492557122,
|
| 11 |
+
"train_runtime": 17933.6413,
|
| 12 |
+
"train_samples_per_second": 4.948,
|
| 13 |
+
"train_steps_per_second": 0.309
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"eval_loss": 0.1913878321647644,
|
| 4 |
+
"eval_na_accuracy": 0.8505154848098755,
|
| 5 |
+
"eval_ordinal_accuracy": 0.6843051910400391,
|
| 6 |
+
"eval_ordinal_mae": 0.41984713077545166,
|
| 7 |
+
"eval_runtime": 173.1798,
|
| 8 |
+
"eval_samples_per_second": 25.84,
|
| 9 |
+
"eval_steps_per_second": 3.234
|
| 10 |
+
}
|
runs/Mar31_17-12-55_ryanserver/events.out.tfevents.1711937747.ryanserver.31714.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc41af4234ec2ac0cfeb657f9fa61ec5019fbe24b50f0a3178e538551af756d5
|
| 3 |
+
size 529
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.18857605492557122,
|
| 4 |
+
"train_runtime": 17933.6413,
|
| 5 |
+
"train_samples_per_second": 4.948,
|
| 6 |
+
"train_steps_per_second": 0.309
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2182 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1913878321647644,
|
| 3 |
+
"best_model_checkpoint": "./ryan03312024_lr_2e-5_wd_001_v2/checkpoint-2800",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 5546,
|
| 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": 0.6066028475761414,
|
| 14 |
+
"learning_rate": 1.9909844933285253e-05,
|
| 15 |
+
"loss": 0.571,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 0.938056468963623,
|
| 21 |
+
"learning_rate": 1.9819689866570503e-05,
|
| 22 |
+
"loss": 0.4613,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03,
|
| 27 |
+
"grad_norm": 1.0178314447402954,
|
| 28 |
+
"learning_rate": 1.9729534799855754e-05,
|
| 29 |
+
"loss": 0.4095,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04,
|
| 34 |
+
"grad_norm": 1.4264367818832397,
|
| 35 |
+
"learning_rate": 1.9639379733141005e-05,
|
| 36 |
+
"loss": 0.4426,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04,
|
| 41 |
+
"eval_loss": 0.37066397070884705,
|
| 42 |
+
"eval_na_accuracy": 0.80756014585495,
|
| 43 |
+
"eval_ordinal_accuracy": 0.3408682346343994,
|
| 44 |
+
"eval_ordinal_mae": 0.8707404732704163,
|
| 45 |
+
"eval_runtime": 343.5137,
|
| 46 |
+
"eval_samples_per_second": 13.027,
|
| 47 |
+
"eval_steps_per_second": 1.63,
|
| 48 |
+
"step": 100
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.05,
|
| 52 |
+
"grad_norm": 0.495741069316864,
|
| 53 |
+
"learning_rate": 1.9549224666426253e-05,
|
| 54 |
+
"loss": 0.3942,
|
| 55 |
+
"step": 125
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.05,
|
| 59 |
+
"grad_norm": 0.6811705827713013,
|
| 60 |
+
"learning_rate": 1.9459069599711507e-05,
|
| 61 |
+
"loss": 0.3667,
|
| 62 |
+
"step": 150
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.06,
|
| 66 |
+
"grad_norm": 1.048833966255188,
|
| 67 |
+
"learning_rate": 1.9368914532996758e-05,
|
| 68 |
+
"loss": 0.3792,
|
| 69 |
+
"step": 175
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.07,
|
| 73 |
+
"grad_norm": 0.517120897769928,
|
| 74 |
+
"learning_rate": 1.9278759466282005e-05,
|
| 75 |
+
"loss": 0.3133,
|
| 76 |
+
"step": 200
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.07,
|
| 80 |
+
"eval_loss": 0.3202913999557495,
|
| 81 |
+
"eval_na_accuracy": 0.7749140858650208,
|
| 82 |
+
"eval_ordinal_accuracy": 0.4300025701522827,
|
| 83 |
+
"eval_ordinal_mae": 0.8544500470161438,
|
| 84 |
+
"eval_runtime": 201.5044,
|
| 85 |
+
"eval_samples_per_second": 22.208,
|
| 86 |
+
"eval_steps_per_second": 2.779,
|
| 87 |
+
"step": 200
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.08,
|
| 91 |
+
"grad_norm": 1.3575235605239868,
|
| 92 |
+
"learning_rate": 1.9188604399567256e-05,
|
| 93 |
+
"loss": 0.3491,
|
| 94 |
+
"step": 225
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.09,
|
| 98 |
+
"grad_norm": 1.0482808351516724,
|
| 99 |
+
"learning_rate": 1.909844933285251e-05,
|
| 100 |
+
"loss": 0.3655,
|
| 101 |
+
"step": 250
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.1,
|
| 105 |
+
"grad_norm": 2.861924409866333,
|
| 106 |
+
"learning_rate": 1.9008294266137758e-05,
|
| 107 |
+
"loss": 0.3141,
|
| 108 |
+
"step": 275
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.11,
|
| 112 |
+
"grad_norm": 1.2723792791366577,
|
| 113 |
+
"learning_rate": 1.891813919942301e-05,
|
| 114 |
+
"loss": 0.3349,
|
| 115 |
+
"step": 300
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.11,
|
| 119 |
+
"eval_loss": 0.2997085452079773,
|
| 120 |
+
"eval_na_accuracy": 0.8419243693351746,
|
| 121 |
+
"eval_ordinal_accuracy": 0.4592858850955963,
|
| 122 |
+
"eval_ordinal_mae": 0.8338660001754761,
|
| 123 |
+
"eval_runtime": 195.4794,
|
| 124 |
+
"eval_samples_per_second": 22.892,
|
| 125 |
+
"eval_steps_per_second": 2.865,
|
| 126 |
+
"step": 300
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.12,
|
| 130 |
+
"grad_norm": 0.6588447690010071,
|
| 131 |
+
"learning_rate": 1.882798413270826e-05,
|
| 132 |
+
"loss": 0.2944,
|
| 133 |
+
"step": 325
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.13,
|
| 137 |
+
"grad_norm": 0.5759406089782715,
|
| 138 |
+
"learning_rate": 1.873782906599351e-05,
|
| 139 |
+
"loss": 0.2992,
|
| 140 |
+
"step": 350
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.14,
|
| 144 |
+
"grad_norm": 0.21856403350830078,
|
| 145 |
+
"learning_rate": 1.864767399927876e-05,
|
| 146 |
+
"loss": 0.305,
|
| 147 |
+
"step": 375
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.14,
|
| 151 |
+
"grad_norm": 1.3187426328659058,
|
| 152 |
+
"learning_rate": 1.8557518932564012e-05,
|
| 153 |
+
"loss": 0.3173,
|
| 154 |
+
"step": 400
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.14,
|
| 158 |
+
"eval_loss": 0.28696730732917786,
|
| 159 |
+
"eval_na_accuracy": 0.8659793734550476,
|
| 160 |
+
"eval_ordinal_accuracy": 0.48189055919647217,
|
| 161 |
+
"eval_ordinal_mae": 0.7992547750473022,
|
| 162 |
+
"eval_runtime": 196.5851,
|
| 163 |
+
"eval_samples_per_second": 22.764,
|
| 164 |
+
"eval_steps_per_second": 2.849,
|
| 165 |
+
"step": 400
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.15,
|
| 169 |
+
"grad_norm": 1.22658109664917,
|
| 170 |
+
"learning_rate": 1.846736386584926e-05,
|
| 171 |
+
"loss": 0.2952,
|
| 172 |
+
"step": 425
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.16,
|
| 176 |
+
"grad_norm": 2.2546815872192383,
|
| 177 |
+
"learning_rate": 1.8377208799134514e-05,
|
| 178 |
+
"loss": 0.3037,
|
| 179 |
+
"step": 450
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.17,
|
| 183 |
+
"grad_norm": 2.8508079051971436,
|
| 184 |
+
"learning_rate": 1.8287053732419765e-05,
|
| 185 |
+
"loss": 0.3114,
|
| 186 |
+
"step": 475
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.18,
|
| 190 |
+
"grad_norm": 1.2727103233337402,
|
| 191 |
+
"learning_rate": 1.8196898665705013e-05,
|
| 192 |
+
"loss": 0.2946,
|
| 193 |
+
"step": 500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.18,
|
| 197 |
+
"eval_loss": 0.2856302559375763,
|
| 198 |
+
"eval_na_accuracy": 0.8934707641601562,
|
| 199 |
+
"eval_ordinal_accuracy": 0.5111739039421082,
|
| 200 |
+
"eval_ordinal_mae": 0.7690269351005554,
|
| 201 |
+
"eval_runtime": 195.0911,
|
| 202 |
+
"eval_samples_per_second": 22.938,
|
| 203 |
+
"eval_steps_per_second": 2.87,
|
| 204 |
+
"step": 500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.19,
|
| 208 |
+
"grad_norm": 1.6682345867156982,
|
| 209 |
+
"learning_rate": 1.8106743598990263e-05,
|
| 210 |
+
"loss": 0.301,
|
| 211 |
+
"step": 525
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.2,
|
| 215 |
+
"grad_norm": 1.0263928174972534,
|
| 216 |
+
"learning_rate": 1.8016588532275518e-05,
|
| 217 |
+
"loss": 0.2746,
|
| 218 |
+
"step": 550
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.21,
|
| 222 |
+
"grad_norm": 0.3150116503238678,
|
| 223 |
+
"learning_rate": 1.7926433465560765e-05,
|
| 224 |
+
"loss": 0.2676,
|
| 225 |
+
"step": 575
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.22,
|
| 229 |
+
"grad_norm": 0.7826778888702393,
|
| 230 |
+
"learning_rate": 1.7836278398846016e-05,
|
| 231 |
+
"loss": 0.3002,
|
| 232 |
+
"step": 600
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.22,
|
| 236 |
+
"eval_loss": 0.2724498510360718,
|
| 237 |
+
"eval_na_accuracy": 0.9209622144699097,
|
| 238 |
+
"eval_ordinal_accuracy": 0.5345491766929626,
|
| 239 |
+
"eval_ordinal_mae": 0.7232748866081238,
|
| 240 |
+
"eval_runtime": 200.1722,
|
| 241 |
+
"eval_samples_per_second": 22.356,
|
| 242 |
+
"eval_steps_per_second": 2.798,
|
| 243 |
+
"step": 600
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.23,
|
| 247 |
+
"grad_norm": 0.7353246808052063,
|
| 248 |
+
"learning_rate": 1.7746123332131267e-05,
|
| 249 |
+
"loss": 0.303,
|
| 250 |
+
"step": 625
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.23,
|
| 254 |
+
"grad_norm": 0.5406638383865356,
|
| 255 |
+
"learning_rate": 1.7655968265416518e-05,
|
| 256 |
+
"loss": 0.2909,
|
| 257 |
+
"step": 650
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.24,
|
| 261 |
+
"grad_norm": 1.443584680557251,
|
| 262 |
+
"learning_rate": 1.756581319870177e-05,
|
| 263 |
+
"loss": 0.2512,
|
| 264 |
+
"step": 675
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.25,
|
| 268 |
+
"grad_norm": 2.028228521347046,
|
| 269 |
+
"learning_rate": 1.747565813198702e-05,
|
| 270 |
+
"loss": 0.2817,
|
| 271 |
+
"step": 700
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.25,
|
| 275 |
+
"eval_loss": 0.26569074392318726,
|
| 276 |
+
"eval_na_accuracy": 0.8625429272651672,
|
| 277 |
+
"eval_ordinal_accuracy": 0.5566401481628418,
|
| 278 |
+
"eval_ordinal_mae": 0.6927691102027893,
|
| 279 |
+
"eval_runtime": 201.4281,
|
| 280 |
+
"eval_samples_per_second": 22.216,
|
| 281 |
+
"eval_steps_per_second": 2.78,
|
| 282 |
+
"step": 700
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.26,
|
| 286 |
+
"grad_norm": 0.40620824694633484,
|
| 287 |
+
"learning_rate": 1.738550306527227e-05,
|
| 288 |
+
"loss": 0.2732,
|
| 289 |
+
"step": 725
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.27,
|
| 293 |
+
"grad_norm": 1.9016238451004028,
|
| 294 |
+
"learning_rate": 1.729534799855752e-05,
|
| 295 |
+
"loss": 0.2552,
|
| 296 |
+
"step": 750
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.28,
|
| 300 |
+
"grad_norm": 2.003732919692993,
|
| 301 |
+
"learning_rate": 1.7205192931842772e-05,
|
| 302 |
+
"loss": 0.2538,
|
| 303 |
+
"step": 775
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.29,
|
| 307 |
+
"grad_norm": 2.751253604888916,
|
| 308 |
+
"learning_rate": 1.711503786512802e-05,
|
| 309 |
+
"loss": 0.2939,
|
| 310 |
+
"step": 800
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.29,
|
| 314 |
+
"eval_loss": 0.25960293412208557,
|
| 315 |
+
"eval_na_accuracy": 0.7920961976051331,
|
| 316 |
+
"eval_ordinal_accuracy": 0.5861803293228149,
|
| 317 |
+
"eval_ordinal_mae": 0.6425440907478333,
|
| 318 |
+
"eval_runtime": 200.1318,
|
| 319 |
+
"eval_samples_per_second": 22.36,
|
| 320 |
+
"eval_steps_per_second": 2.798,
|
| 321 |
+
"step": 800
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.3,
|
| 325 |
+
"grad_norm": 2.241727113723755,
|
| 326 |
+
"learning_rate": 1.702488279841327e-05,
|
| 327 |
+
"loss": 0.2659,
|
| 328 |
+
"step": 825
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.31,
|
| 332 |
+
"grad_norm": 0.8341929912567139,
|
| 333 |
+
"learning_rate": 1.6938333934367113e-05,
|
| 334 |
+
"loss": 0.2654,
|
| 335 |
+
"step": 850
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.32,
|
| 339 |
+
"grad_norm": 1.8499948978424072,
|
| 340 |
+
"learning_rate": 1.6848178867652364e-05,
|
| 341 |
+
"loss": 0.2774,
|
| 342 |
+
"step": 875
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.32,
|
| 346 |
+
"grad_norm": 1.8184664249420166,
|
| 347 |
+
"learning_rate": 1.6758023800937615e-05,
|
| 348 |
+
"loss": 0.2525,
|
| 349 |
+
"step": 900
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.32,
|
| 353 |
+
"eval_loss": 0.24585017561912537,
|
| 354 |
+
"eval_na_accuracy": 0.8264604806900024,
|
| 355 |
+
"eval_ordinal_accuracy": 0.6046750545501709,
|
| 356 |
+
"eval_ordinal_mae": 0.6053154468536377,
|
| 357 |
+
"eval_runtime": 203.517,
|
| 358 |
+
"eval_samples_per_second": 21.988,
|
| 359 |
+
"eval_steps_per_second": 2.752,
|
| 360 |
+
"step": 900
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.33,
|
| 364 |
+
"grad_norm": 1.131605625152588,
|
| 365 |
+
"learning_rate": 1.6667868734222866e-05,
|
| 366 |
+
"loss": 0.2701,
|
| 367 |
+
"step": 925
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.34,
|
| 371 |
+
"grad_norm": 2.05479097366333,
|
| 372 |
+
"learning_rate": 1.6577713667508117e-05,
|
| 373 |
+
"loss": 0.2956,
|
| 374 |
+
"step": 950
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.35,
|
| 378 |
+
"grad_norm": 1.3499815464019775,
|
| 379 |
+
"learning_rate": 1.6487558600793364e-05,
|
| 380 |
+
"loss": 0.2446,
|
| 381 |
+
"step": 975
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.36,
|
| 385 |
+
"grad_norm": 0.7456974983215332,
|
| 386 |
+
"learning_rate": 1.639740353407862e-05,
|
| 387 |
+
"loss": 0.2163,
|
| 388 |
+
"step": 1000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.36,
|
| 392 |
+
"eval_loss": 0.2399808019399643,
|
| 393 |
+
"eval_na_accuracy": 0.8109965920448303,
|
| 394 |
+
"eval_ordinal_accuracy": 0.624454140663147,
|
| 395 |
+
"eval_ordinal_mae": 0.5777344703674316,
|
| 396 |
+
"eval_runtime": 217.2069,
|
| 397 |
+
"eval_samples_per_second": 20.602,
|
| 398 |
+
"eval_steps_per_second": 2.578,
|
| 399 |
+
"step": 1000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.37,
|
| 403 |
+
"grad_norm": 1.0125823020935059,
|
| 404 |
+
"learning_rate": 1.6307248467363866e-05,
|
| 405 |
+
"loss": 0.2168,
|
| 406 |
+
"step": 1025
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.38,
|
| 410 |
+
"grad_norm": 1.9579060077667236,
|
| 411 |
+
"learning_rate": 1.6217093400649117e-05,
|
| 412 |
+
"loss": 0.2364,
|
| 413 |
+
"step": 1050
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.39,
|
| 417 |
+
"grad_norm": 2.975767135620117,
|
| 418 |
+
"learning_rate": 1.6126938333934368e-05,
|
| 419 |
+
"loss": 0.2716,
|
| 420 |
+
"step": 1075
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.4,
|
| 424 |
+
"grad_norm": 2.9766530990600586,
|
| 425 |
+
"learning_rate": 1.603678326721962e-05,
|
| 426 |
+
"loss": 0.2181,
|
| 427 |
+
"step": 1100
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.4,
|
| 431 |
+
"eval_loss": 0.23391272127628326,
|
| 432 |
+
"eval_na_accuracy": 0.876288652420044,
|
| 433 |
+
"eval_ordinal_accuracy": 0.6023632287979126,
|
| 434 |
+
"eval_ordinal_mae": 0.5429691672325134,
|
| 435 |
+
"eval_runtime": 209.4151,
|
| 436 |
+
"eval_samples_per_second": 21.369,
|
| 437 |
+
"eval_steps_per_second": 2.674,
|
| 438 |
+
"step": 1100
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.41,
|
| 442 |
+
"grad_norm": 0.6088442802429199,
|
| 443 |
+
"learning_rate": 1.594662820050487e-05,
|
| 444 |
+
"loss": 0.2128,
|
| 445 |
+
"step": 1125
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.41,
|
| 449 |
+
"grad_norm": 0.3461017906665802,
|
| 450 |
+
"learning_rate": 1.585647313379012e-05,
|
| 451 |
+
"loss": 0.2145,
|
| 452 |
+
"step": 1150
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.42,
|
| 456 |
+
"grad_norm": 1.341361165046692,
|
| 457 |
+
"learning_rate": 1.576631806707537e-05,
|
| 458 |
+
"loss": 0.2554,
|
| 459 |
+
"step": 1175
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.43,
|
| 463 |
+
"grad_norm": 0.5675370693206787,
|
| 464 |
+
"learning_rate": 1.5676163000360622e-05,
|
| 465 |
+
"loss": 0.1949,
|
| 466 |
+
"step": 1200
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.43,
|
| 470 |
+
"eval_loss": 0.23308563232421875,
|
| 471 |
+
"eval_na_accuracy": 0.7955326437950134,
|
| 472 |
+
"eval_ordinal_accuracy": 0.6285640597343445,
|
| 473 |
+
"eval_ordinal_mae": 0.5328579545021057,
|
| 474 |
+
"eval_runtime": 209.3844,
|
| 475 |
+
"eval_samples_per_second": 21.372,
|
| 476 |
+
"eval_steps_per_second": 2.675,
|
| 477 |
+
"step": 1200
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.44,
|
| 481 |
+
"grad_norm": 2.334373712539673,
|
| 482 |
+
"learning_rate": 1.5586007933645873e-05,
|
| 483 |
+
"loss": 0.2739,
|
| 484 |
+
"step": 1225
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.45,
|
| 488 |
+
"grad_norm": 0.5421320199966431,
|
| 489 |
+
"learning_rate": 1.5495852866931124e-05,
|
| 490 |
+
"loss": 0.2294,
|
| 491 |
+
"step": 1250
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.46,
|
| 495 |
+
"grad_norm": 2.599592924118042,
|
| 496 |
+
"learning_rate": 1.540569780021637e-05,
|
| 497 |
+
"loss": 0.255,
|
| 498 |
+
"step": 1275
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.47,
|
| 502 |
+
"grad_norm": 0.6286499500274658,
|
| 503 |
+
"learning_rate": 1.5315542733501626e-05,
|
| 504 |
+
"loss": 0.214,
|
| 505 |
+
"step": 1300
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.47,
|
| 509 |
+
"eval_loss": 0.24238631129264832,
|
| 510 |
+
"eval_na_accuracy": 0.7628865838050842,
|
| 511 |
+
"eval_ordinal_accuracy": 0.6182892322540283,
|
| 512 |
+
"eval_ordinal_mae": 0.5243560671806335,
|
| 513 |
+
"eval_runtime": 212.87,
|
| 514 |
+
"eval_samples_per_second": 21.022,
|
| 515 |
+
"eval_steps_per_second": 2.631,
|
| 516 |
+
"step": 1300
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 0.48,
|
| 520 |
+
"grad_norm": 1.9406815767288208,
|
| 521 |
+
"learning_rate": 1.5225387666786875e-05,
|
| 522 |
+
"loss": 0.2225,
|
| 523 |
+
"step": 1325
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 0.49,
|
| 527 |
+
"grad_norm": 3.733520269393921,
|
| 528 |
+
"learning_rate": 1.5135232600072126e-05,
|
| 529 |
+
"loss": 0.1918,
|
| 530 |
+
"step": 1350
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.5,
|
| 534 |
+
"grad_norm": 0.8792308568954468,
|
| 535 |
+
"learning_rate": 1.5045077533357375e-05,
|
| 536 |
+
"loss": 0.2179,
|
| 537 |
+
"step": 1375
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.5,
|
| 541 |
+
"grad_norm": 1.1207715272903442,
|
| 542 |
+
"learning_rate": 1.4954922466642628e-05,
|
| 543 |
+
"loss": 0.27,
|
| 544 |
+
"step": 1400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.5,
|
| 548 |
+
"eval_loss": 0.22981737554073334,
|
| 549 |
+
"eval_na_accuracy": 0.7869415879249573,
|
| 550 |
+
"eval_ordinal_accuracy": 0.6367839574813843,
|
| 551 |
+
"eval_ordinal_mae": 0.4994910955429077,
|
| 552 |
+
"eval_runtime": 212.2299,
|
| 553 |
+
"eval_samples_per_second": 21.086,
|
| 554 |
+
"eval_steps_per_second": 2.639,
|
| 555 |
+
"step": 1400
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.51,
|
| 559 |
+
"grad_norm": 2.47230863571167,
|
| 560 |
+
"learning_rate": 1.4864767399927877e-05,
|
| 561 |
+
"loss": 0.2348,
|
| 562 |
+
"step": 1425
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.52,
|
| 566 |
+
"grad_norm": 1.015068769454956,
|
| 567 |
+
"learning_rate": 1.4774612333213128e-05,
|
| 568 |
+
"loss": 0.2593,
|
| 569 |
+
"step": 1450
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.53,
|
| 573 |
+
"grad_norm": 1.7957515716552734,
|
| 574 |
+
"learning_rate": 1.4684457266498377e-05,
|
| 575 |
+
"loss": 0.2049,
|
| 576 |
+
"step": 1475
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.54,
|
| 580 |
+
"grad_norm": 1.9824727773666382,
|
| 581 |
+
"learning_rate": 1.459430219978363e-05,
|
| 582 |
+
"loss": 0.2117,
|
| 583 |
+
"step": 1500
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.54,
|
| 587 |
+
"eval_loss": 0.2300846129655838,
|
| 588 |
+
"eval_na_accuracy": 0.7783505320549011,
|
| 589 |
+
"eval_ordinal_accuracy": 0.6473156809806824,
|
| 590 |
+
"eval_ordinal_mae": 0.4949534237384796,
|
| 591 |
+
"eval_runtime": 210.619,
|
| 592 |
+
"eval_samples_per_second": 21.247,
|
| 593 |
+
"eval_steps_per_second": 2.659,
|
| 594 |
+
"step": 1500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.55,
|
| 598 |
+
"grad_norm": 1.1805070638656616,
|
| 599 |
+
"learning_rate": 1.450414713306888e-05,
|
| 600 |
+
"loss": 0.189,
|
| 601 |
+
"step": 1525
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.56,
|
| 605 |
+
"grad_norm": 1.2057225704193115,
|
| 606 |
+
"learning_rate": 1.441399206635413e-05,
|
| 607 |
+
"loss": 0.2254,
|
| 608 |
+
"step": 1550
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.57,
|
| 612 |
+
"grad_norm": 3.1214776039123535,
|
| 613 |
+
"learning_rate": 1.432383699963938e-05,
|
| 614 |
+
"loss": 0.2286,
|
| 615 |
+
"step": 1575
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.58,
|
| 619 |
+
"grad_norm": 2.340480089187622,
|
| 620 |
+
"learning_rate": 1.4233681932924633e-05,
|
| 621 |
+
"loss": 0.2038,
|
| 622 |
+
"step": 1600
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.58,
|
| 626 |
+
"eval_loss": 0.21561825275421143,
|
| 627 |
+
"eval_na_accuracy": 0.8367697596549988,
|
| 628 |
+
"eval_ordinal_accuracy": 0.6550218462944031,
|
| 629 |
+
"eval_ordinal_mae": 0.48989665508270264,
|
| 630 |
+
"eval_runtime": 207.7331,
|
| 631 |
+
"eval_samples_per_second": 21.542,
|
| 632 |
+
"eval_steps_per_second": 2.696,
|
| 633 |
+
"step": 1600
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.59,
|
| 637 |
+
"grad_norm": 2.1810505390167236,
|
| 638 |
+
"learning_rate": 1.4143526866209882e-05,
|
| 639 |
+
"loss": 0.2223,
|
| 640 |
+
"step": 1625
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.6,
|
| 644 |
+
"grad_norm": 2.7431559562683105,
|
| 645 |
+
"learning_rate": 1.4053371799495133e-05,
|
| 646 |
+
"loss": 0.2644,
|
| 647 |
+
"step": 1650
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6,
|
| 651 |
+
"grad_norm": 0.8831206560134888,
|
| 652 |
+
"learning_rate": 1.3963216732780382e-05,
|
| 653 |
+
"loss": 0.2589,
|
| 654 |
+
"step": 1675
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.61,
|
| 658 |
+
"grad_norm": 2.4183125495910645,
|
| 659 |
+
"learning_rate": 1.3873061666065635e-05,
|
| 660 |
+
"loss": 0.1974,
|
| 661 |
+
"step": 1700
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.61,
|
| 665 |
+
"eval_loss": 0.22116971015930176,
|
| 666 |
+
"eval_na_accuracy": 0.8281787037849426,
|
| 667 |
+
"eval_ordinal_accuracy": 0.6347290277481079,
|
| 668 |
+
"eval_ordinal_mae": 0.4638911485671997,
|
| 669 |
+
"eval_runtime": 207.3318,
|
| 670 |
+
"eval_samples_per_second": 21.584,
|
| 671 |
+
"eval_steps_per_second": 2.701,
|
| 672 |
+
"step": 1700
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.62,
|
| 676 |
+
"grad_norm": 1.318037986755371,
|
| 677 |
+
"learning_rate": 1.3782906599350886e-05,
|
| 678 |
+
"loss": 0.2456,
|
| 679 |
+
"step": 1725
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 0.63,
|
| 683 |
+
"grad_norm": 3.9031999111175537,
|
| 684 |
+
"learning_rate": 1.3692751532636135e-05,
|
| 685 |
+
"loss": 0.2208,
|
| 686 |
+
"step": 1750
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.64,
|
| 690 |
+
"grad_norm": 0.9724792838096619,
|
| 691 |
+
"learning_rate": 1.3602596465921386e-05,
|
| 692 |
+
"loss": 0.1694,
|
| 693 |
+
"step": 1775
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 0.65,
|
| 697 |
+
"grad_norm": 0.8028815984725952,
|
| 698 |
+
"learning_rate": 1.3512441399206637e-05,
|
| 699 |
+
"loss": 0.1916,
|
| 700 |
+
"step": 1800
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 0.65,
|
| 704 |
+
"eval_loss": 0.2150656282901764,
|
| 705 |
+
"eval_na_accuracy": 0.8797250986099243,
|
| 706 |
+
"eval_ordinal_accuracy": 0.6439763903617859,
|
| 707 |
+
"eval_ordinal_mae": 0.4789562523365021,
|
| 708 |
+
"eval_runtime": 211.6606,
|
| 709 |
+
"eval_samples_per_second": 21.142,
|
| 710 |
+
"eval_steps_per_second": 2.646,
|
| 711 |
+
"step": 1800
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.66,
|
| 715 |
+
"grad_norm": 5.7413434982299805,
|
| 716 |
+
"learning_rate": 1.3422286332491887e-05,
|
| 717 |
+
"loss": 0.2097,
|
| 718 |
+
"step": 1825
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.67,
|
| 722 |
+
"grad_norm": 0.586044192314148,
|
| 723 |
+
"learning_rate": 1.3332131265777137e-05,
|
| 724 |
+
"loss": 0.1926,
|
| 725 |
+
"step": 1850
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.68,
|
| 729 |
+
"grad_norm": 3.0166969299316406,
|
| 730 |
+
"learning_rate": 1.3241976199062388e-05,
|
| 731 |
+
"loss": 0.1901,
|
| 732 |
+
"step": 1875
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.69,
|
| 736 |
+
"grad_norm": 0.5749481916427612,
|
| 737 |
+
"learning_rate": 1.315182113234764e-05,
|
| 738 |
+
"loss": 0.1921,
|
| 739 |
+
"step": 1900
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.69,
|
| 743 |
+
"eval_loss": 0.20500315725803375,
|
| 744 |
+
"eval_na_accuracy": 0.8728522062301636,
|
| 745 |
+
"eval_ordinal_accuracy": 0.6609298586845398,
|
| 746 |
+
"eval_ordinal_mae": 0.4614226222038269,
|
| 747 |
+
"eval_runtime": 179.4835,
|
| 748 |
+
"eval_samples_per_second": 24.933,
|
| 749 |
+
"eval_steps_per_second": 3.12,
|
| 750 |
+
"step": 1900
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.69,
|
| 754 |
+
"grad_norm": 1.0241420269012451,
|
| 755 |
+
"learning_rate": 1.306166606563289e-05,
|
| 756 |
+
"loss": 0.2312,
|
| 757 |
+
"step": 1925
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.7,
|
| 761 |
+
"grad_norm": 2.6365127563476562,
|
| 762 |
+
"learning_rate": 1.2975117201586732e-05,
|
| 763 |
+
"loss": 0.2358,
|
| 764 |
+
"step": 1950
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.71,
|
| 768 |
+
"grad_norm": 2.4451234340667725,
|
| 769 |
+
"learning_rate": 1.2884962134871981e-05,
|
| 770 |
+
"loss": 0.2349,
|
| 771 |
+
"step": 1975
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.72,
|
| 775 |
+
"grad_norm": 0.9995049238204956,
|
| 776 |
+
"learning_rate": 1.2794807068157232e-05,
|
| 777 |
+
"loss": 0.1936,
|
| 778 |
+
"step": 2000
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.72,
|
| 782 |
+
"eval_loss": 0.20607997477054596,
|
| 783 |
+
"eval_na_accuracy": 0.8573883175849915,
|
| 784 |
+
"eval_ordinal_accuracy": 0.6496275663375854,
|
| 785 |
+
"eval_ordinal_mae": 0.4565940797328949,
|
| 786 |
+
"eval_runtime": 175.1471,
|
| 787 |
+
"eval_samples_per_second": 25.55,
|
| 788 |
+
"eval_steps_per_second": 3.197,
|
| 789 |
+
"step": 2000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.73,
|
| 793 |
+
"grad_norm": 5.394123554229736,
|
| 794 |
+
"learning_rate": 1.2704652001442481e-05,
|
| 795 |
+
"loss": 0.2187,
|
| 796 |
+
"step": 2025
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.74,
|
| 800 |
+
"grad_norm": 2.7290520668029785,
|
| 801 |
+
"learning_rate": 1.2614496934727734e-05,
|
| 802 |
+
"loss": 0.2144,
|
| 803 |
+
"step": 2050
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.75,
|
| 807 |
+
"grad_norm": 2.8023579120635986,
|
| 808 |
+
"learning_rate": 1.2524341868012983e-05,
|
| 809 |
+
"loss": 0.1749,
|
| 810 |
+
"step": 2075
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.76,
|
| 814 |
+
"grad_norm": 2.040255308151245,
|
| 815 |
+
"learning_rate": 1.2434186801298234e-05,
|
| 816 |
+
"loss": 0.1939,
|
| 817 |
+
"step": 2100
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.76,
|
| 821 |
+
"eval_loss": 0.22944368422031403,
|
| 822 |
+
"eval_na_accuracy": 0.9089347124099731,
|
| 823 |
+
"eval_ordinal_accuracy": 0.6362702250480652,
|
| 824 |
+
"eval_ordinal_mae": 0.4656565487384796,
|
| 825 |
+
"eval_runtime": 176.4408,
|
| 826 |
+
"eval_samples_per_second": 25.363,
|
| 827 |
+
"eval_steps_per_second": 3.174,
|
| 828 |
+
"step": 2100
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.77,
|
| 832 |
+
"grad_norm": 1.4762495756149292,
|
| 833 |
+
"learning_rate": 1.2344031734583483e-05,
|
| 834 |
+
"loss": 0.2211,
|
| 835 |
+
"step": 2125
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.78,
|
| 839 |
+
"grad_norm": 3.6306161880493164,
|
| 840 |
+
"learning_rate": 1.2253876667868735e-05,
|
| 841 |
+
"loss": 0.2548,
|
| 842 |
+
"step": 2150
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.78,
|
| 846 |
+
"grad_norm": 1.6580469608306885,
|
| 847 |
+
"learning_rate": 1.2163721601153986e-05,
|
| 848 |
+
"loss": 0.211,
|
| 849 |
+
"step": 2175
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.79,
|
| 853 |
+
"grad_norm": 2.7241005897521973,
|
| 854 |
+
"learning_rate": 1.2073566534439236e-05,
|
| 855 |
+
"loss": 0.257,
|
| 856 |
+
"step": 2200
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.79,
|
| 860 |
+
"eval_loss": 0.20536109805107117,
|
| 861 |
+
"eval_na_accuracy": 0.8608247637748718,
|
| 862 |
+
"eval_ordinal_accuracy": 0.6527100205421448,
|
| 863 |
+
"eval_ordinal_mae": 0.45671576261520386,
|
| 864 |
+
"eval_runtime": 176.5474,
|
| 865 |
+
"eval_samples_per_second": 25.347,
|
| 866 |
+
"eval_steps_per_second": 3.172,
|
| 867 |
+
"step": 2200
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.8,
|
| 871 |
+
"grad_norm": 2.5323662757873535,
|
| 872 |
+
"learning_rate": 1.1983411467724486e-05,
|
| 873 |
+
"loss": 0.176,
|
| 874 |
+
"step": 2225
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.81,
|
| 878 |
+
"grad_norm": 5.30660343170166,
|
| 879 |
+
"learning_rate": 1.1893256401009739e-05,
|
| 880 |
+
"loss": 0.195,
|
| 881 |
+
"step": 2250
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.82,
|
| 885 |
+
"grad_norm": 3.7983808517456055,
|
| 886 |
+
"learning_rate": 1.1803101334294988e-05,
|
| 887 |
+
"loss": 0.1746,
|
| 888 |
+
"step": 2275
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 0.83,
|
| 892 |
+
"grad_norm": 0.6054279804229736,
|
| 893 |
+
"learning_rate": 1.1712946267580239e-05,
|
| 894 |
+
"loss": 0.2236,
|
| 895 |
+
"step": 2300
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.83,
|
| 899 |
+
"eval_loss": 0.2043941468000412,
|
| 900 |
+
"eval_na_accuracy": 0.876288652420044,
|
| 901 |
+
"eval_ordinal_accuracy": 0.6640123128890991,
|
| 902 |
+
"eval_ordinal_mae": 0.45420947670936584,
|
| 903 |
+
"eval_runtime": 176.816,
|
| 904 |
+
"eval_samples_per_second": 25.309,
|
| 905 |
+
"eval_steps_per_second": 3.167,
|
| 906 |
+
"step": 2300
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.84,
|
| 910 |
+
"grad_norm": 1.1567206382751465,
|
| 911 |
+
"learning_rate": 1.1622791200865488e-05,
|
| 912 |
+
"loss": 0.2479,
|
| 913 |
+
"step": 2325
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.85,
|
| 917 |
+
"grad_norm": 5.52665376663208,
|
| 918 |
+
"learning_rate": 1.1532636134150741e-05,
|
| 919 |
+
"loss": 0.2271,
|
| 920 |
+
"step": 2350
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.86,
|
| 924 |
+
"grad_norm": 2.9272255897521973,
|
| 925 |
+
"learning_rate": 1.1442481067435992e-05,
|
| 926 |
+
"loss": 0.2303,
|
| 927 |
+
"step": 2375
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.87,
|
| 931 |
+
"grad_norm": 3.266409397125244,
|
| 932 |
+
"learning_rate": 1.1352326000721241e-05,
|
| 933 |
+
"loss": 0.1925,
|
| 934 |
+
"step": 2400
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.87,
|
| 938 |
+
"eval_loss": 0.20849426090717316,
|
| 939 |
+
"eval_na_accuracy": 0.80756014585495,
|
| 940 |
+
"eval_ordinal_accuracy": 0.6886719465255737,
|
| 941 |
+
"eval_ordinal_mae": 0.44630834460258484,
|
| 942 |
+
"eval_runtime": 175.9296,
|
| 943 |
+
"eval_samples_per_second": 25.436,
|
| 944 |
+
"eval_steps_per_second": 3.183,
|
| 945 |
+
"step": 2400
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.87,
|
| 949 |
+
"grad_norm": 2.3057689666748047,
|
| 950 |
+
"learning_rate": 1.1262170934006492e-05,
|
| 951 |
+
"loss": 0.2049,
|
| 952 |
+
"step": 2425
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 0.88,
|
| 956 |
+
"grad_norm": 3.8032755851745605,
|
| 957 |
+
"learning_rate": 1.1172015867291743e-05,
|
| 958 |
+
"loss": 0.1909,
|
| 959 |
+
"step": 2450
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 0.89,
|
| 963 |
+
"grad_norm": 0.4314826428890228,
|
| 964 |
+
"learning_rate": 1.1081860800576994e-05,
|
| 965 |
+
"loss": 0.1825,
|
| 966 |
+
"step": 2475
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 0.9,
|
| 970 |
+
"grad_norm": 3.1127185821533203,
|
| 971 |
+
"learning_rate": 1.0991705733862243e-05,
|
| 972 |
+
"loss": 0.1657,
|
| 973 |
+
"step": 2500
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 0.9,
|
| 977 |
+
"eval_loss": 0.2033829689025879,
|
| 978 |
+
"eval_na_accuracy": 0.8522336483001709,
|
| 979 |
+
"eval_ordinal_accuracy": 0.6768559217453003,
|
| 980 |
+
"eval_ordinal_mae": 0.4391731023788452,
|
| 981 |
+
"eval_runtime": 174.624,
|
| 982 |
+
"eval_samples_per_second": 25.626,
|
| 983 |
+
"eval_steps_per_second": 3.207,
|
| 984 |
+
"step": 2500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.91,
|
| 988 |
+
"grad_norm": 4.6838459968566895,
|
| 989 |
+
"learning_rate": 1.0901550667147494e-05,
|
| 990 |
+
"loss": 0.2397,
|
| 991 |
+
"step": 2525
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.92,
|
| 995 |
+
"grad_norm": 2.440084218978882,
|
| 996 |
+
"learning_rate": 1.0811395600432746e-05,
|
| 997 |
+
"loss": 0.1921,
|
| 998 |
+
"step": 2550
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.93,
|
| 1002 |
+
"grad_norm": 0.5833593606948853,
|
| 1003 |
+
"learning_rate": 1.0721240533717995e-05,
|
| 1004 |
+
"loss": 0.188,
|
| 1005 |
+
"step": 2575
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.94,
|
| 1009 |
+
"grad_norm": 2.1624417304992676,
|
| 1010 |
+
"learning_rate": 1.0631085467003246e-05,
|
| 1011 |
+
"loss": 0.1723,
|
| 1012 |
+
"step": 2600
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 0.94,
|
| 1016 |
+
"eval_loss": 0.1956964135169983,
|
| 1017 |
+
"eval_na_accuracy": 0.838487982749939,
|
| 1018 |
+
"eval_ordinal_accuracy": 0.6755715608596802,
|
| 1019 |
+
"eval_ordinal_mae": 0.42565402388572693,
|
| 1020 |
+
"eval_runtime": 175.3137,
|
| 1021 |
+
"eval_samples_per_second": 25.526,
|
| 1022 |
+
"eval_steps_per_second": 3.194,
|
| 1023 |
+
"step": 2600
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.95,
|
| 1027 |
+
"grad_norm": 3.9376819133758545,
|
| 1028 |
+
"learning_rate": 1.0540930400288496e-05,
|
| 1029 |
+
"loss": 0.1889,
|
| 1030 |
+
"step": 2625
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.96,
|
| 1034 |
+
"grad_norm": 2.2914035320281982,
|
| 1035 |
+
"learning_rate": 1.0450775333573748e-05,
|
| 1036 |
+
"loss": 0.2078,
|
| 1037 |
+
"step": 2650
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.96,
|
| 1041 |
+
"grad_norm": 1.2271968126296997,
|
| 1042 |
+
"learning_rate": 1.0360620266858999e-05,
|
| 1043 |
+
"loss": 0.2247,
|
| 1044 |
+
"step": 2675
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.97,
|
| 1048 |
+
"grad_norm": 2.135350227355957,
|
| 1049 |
+
"learning_rate": 1.0270465200144248e-05,
|
| 1050 |
+
"loss": 0.2279,
|
| 1051 |
+
"step": 2700
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.97,
|
| 1055 |
+
"eval_loss": 0.19462928175926208,
|
| 1056 |
+
"eval_na_accuracy": 0.8642611503601074,
|
| 1057 |
+
"eval_ordinal_accuracy": 0.6740303039550781,
|
| 1058 |
+
"eval_ordinal_mae": 0.4287233054637909,
|
| 1059 |
+
"eval_runtime": 175.4222,
|
| 1060 |
+
"eval_samples_per_second": 25.51,
|
| 1061 |
+
"eval_steps_per_second": 3.192,
|
| 1062 |
+
"step": 2700
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 0.98,
|
| 1066 |
+
"grad_norm": 0.897443413734436,
|
| 1067 |
+
"learning_rate": 1.0180310133429499e-05,
|
| 1068 |
+
"loss": 0.1672,
|
| 1069 |
+
"step": 2725
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 0.99,
|
| 1073 |
+
"grad_norm": 1.9568222761154175,
|
| 1074 |
+
"learning_rate": 1.009015506671475e-05,
|
| 1075 |
+
"loss": 0.1597,
|
| 1076 |
+
"step": 2750
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.0,
|
| 1080 |
+
"grad_norm": 1.222680926322937,
|
| 1081 |
+
"learning_rate": 1e-05,
|
| 1082 |
+
"loss": 0.2093,
|
| 1083 |
+
"step": 2775
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 1.01,
|
| 1087 |
+
"grad_norm": 1.0666455030441284,
|
| 1088 |
+
"learning_rate": 9.909844933285252e-06,
|
| 1089 |
+
"loss": 0.1421,
|
| 1090 |
+
"step": 2800
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.01,
|
| 1094 |
+
"eval_loss": 0.1913878321647644,
|
| 1095 |
+
"eval_na_accuracy": 0.8505154848098755,
|
| 1096 |
+
"eval_ordinal_accuracy": 0.6843051910400391,
|
| 1097 |
+
"eval_ordinal_mae": 0.41984713077545166,
|
| 1098 |
+
"eval_runtime": 177.3377,
|
| 1099 |
+
"eval_samples_per_second": 25.234,
|
| 1100 |
+
"eval_steps_per_second": 3.158,
|
| 1101 |
+
"step": 2800
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.02,
|
| 1105 |
+
"grad_norm": 0.9463332295417786,
|
| 1106 |
+
"learning_rate": 9.819689866570503e-06,
|
| 1107 |
+
"loss": 0.1563,
|
| 1108 |
+
"step": 2825
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.03,
|
| 1112 |
+
"grad_norm": 0.9075057506561279,
|
| 1113 |
+
"learning_rate": 9.729534799855753e-06,
|
| 1114 |
+
"loss": 0.1411,
|
| 1115 |
+
"step": 2850
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.04,
|
| 1119 |
+
"grad_norm": 3.121269702911377,
|
| 1120 |
+
"learning_rate": 9.639379733141003e-06,
|
| 1121 |
+
"loss": 0.1368,
|
| 1122 |
+
"step": 2875
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.05,
|
| 1126 |
+
"grad_norm": 0.5694425702095032,
|
| 1127 |
+
"learning_rate": 9.549224666426255e-06,
|
| 1128 |
+
"loss": 0.1116,
|
| 1129 |
+
"step": 2900
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.05,
|
| 1133 |
+
"eval_loss": 0.20190554857254028,
|
| 1134 |
+
"eval_na_accuracy": 0.8230240345001221,
|
| 1135 |
+
"eval_ordinal_accuracy": 0.6704341173171997,
|
| 1136 |
+
"eval_ordinal_mae": 0.4214249551296234,
|
| 1137 |
+
"eval_runtime": 175.5908,
|
| 1138 |
+
"eval_samples_per_second": 25.485,
|
| 1139 |
+
"eval_steps_per_second": 3.189,
|
| 1140 |
+
"step": 2900
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.05,
|
| 1144 |
+
"grad_norm": 4.727107048034668,
|
| 1145 |
+
"learning_rate": 9.459069599711504e-06,
|
| 1146 |
+
"loss": 0.1485,
|
| 1147 |
+
"step": 2925
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 1.06,
|
| 1151 |
+
"grad_norm": 0.6293134689331055,
|
| 1152 |
+
"learning_rate": 9.368914532996755e-06,
|
| 1153 |
+
"loss": 0.116,
|
| 1154 |
+
"step": 2950
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 1.07,
|
| 1158 |
+
"grad_norm": 0.7508574724197388,
|
| 1159 |
+
"learning_rate": 9.278759466282006e-06,
|
| 1160 |
+
"loss": 0.1656,
|
| 1161 |
+
"step": 2975
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 1.08,
|
| 1165 |
+
"grad_norm": 0.7475106120109558,
|
| 1166 |
+
"learning_rate": 9.188604399567257e-06,
|
| 1167 |
+
"loss": 0.1194,
|
| 1168 |
+
"step": 3000
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.08,
|
| 1172 |
+
"eval_loss": 0.19536536931991577,
|
| 1173 |
+
"eval_na_accuracy": 0.8367697596549988,
|
| 1174 |
+
"eval_ordinal_accuracy": 0.6807089447975159,
|
| 1175 |
+
"eval_ordinal_mae": 0.41778308153152466,
|
| 1176 |
+
"eval_runtime": 175.6051,
|
| 1177 |
+
"eval_samples_per_second": 25.483,
|
| 1178 |
+
"eval_steps_per_second": 3.189,
|
| 1179 |
+
"step": 3000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.09,
|
| 1183 |
+
"grad_norm": 0.6796592473983765,
|
| 1184 |
+
"learning_rate": 9.098449332852506e-06,
|
| 1185 |
+
"loss": 0.1483,
|
| 1186 |
+
"step": 3025
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.1,
|
| 1190 |
+
"grad_norm": 1.7480669021606445,
|
| 1191 |
+
"learning_rate": 9.008294266137759e-06,
|
| 1192 |
+
"loss": 0.1083,
|
| 1193 |
+
"step": 3050
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.11,
|
| 1197 |
+
"grad_norm": 0.7591541409492493,
|
| 1198 |
+
"learning_rate": 8.918139199423008e-06,
|
| 1199 |
+
"loss": 0.1463,
|
| 1200 |
+
"step": 3075
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.12,
|
| 1204 |
+
"grad_norm": 0.8966888189315796,
|
| 1205 |
+
"learning_rate": 8.827984132708259e-06,
|
| 1206 |
+
"loss": 0.1312,
|
| 1207 |
+
"step": 3100
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.12,
|
| 1211 |
+
"eval_loss": 0.19300740957260132,
|
| 1212 |
+
"eval_na_accuracy": 0.8591065406799316,
|
| 1213 |
+
"eval_ordinal_accuracy": 0.6873876452445984,
|
| 1214 |
+
"eval_ordinal_mae": 0.4165719449520111,
|
| 1215 |
+
"eval_runtime": 179.7716,
|
| 1216 |
+
"eval_samples_per_second": 24.893,
|
| 1217 |
+
"eval_steps_per_second": 3.115,
|
| 1218 |
+
"step": 3100
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 1.13,
|
| 1222 |
+
"grad_norm": 1.1078789234161377,
|
| 1223 |
+
"learning_rate": 8.73782906599351e-06,
|
| 1224 |
+
"loss": 0.154,
|
| 1225 |
+
"step": 3125
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 1.14,
|
| 1229 |
+
"grad_norm": 4.303710460662842,
|
| 1230 |
+
"learning_rate": 8.64767399927876e-06,
|
| 1231 |
+
"loss": 0.1328,
|
| 1232 |
+
"step": 3150
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.14,
|
| 1236 |
+
"grad_norm": 0.9769502282142639,
|
| 1237 |
+
"learning_rate": 8.55751893256401e-06,
|
| 1238 |
+
"loss": 0.1253,
|
| 1239 |
+
"step": 3175
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 1.15,
|
| 1243 |
+
"grad_norm": 2.734457492828369,
|
| 1244 |
+
"learning_rate": 8.467363865849262e-06,
|
| 1245 |
+
"loss": 0.1836,
|
| 1246 |
+
"step": 3200
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 1.15,
|
| 1250 |
+
"eval_loss": 0.19890539348125458,
|
| 1251 |
+
"eval_na_accuracy": 0.8642611503601074,
|
| 1252 |
+
"eval_ordinal_accuracy": 0.6794245839118958,
|
| 1253 |
+
"eval_ordinal_mae": 0.4106903374195099,
|
| 1254 |
+
"eval_runtime": 173.0922,
|
| 1255 |
+
"eval_samples_per_second": 25.853,
|
| 1256 |
+
"eval_steps_per_second": 3.235,
|
| 1257 |
+
"step": 3200
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 1.16,
|
| 1261 |
+
"grad_norm": 1.8836966753005981,
|
| 1262 |
+
"learning_rate": 8.377208799134512e-06,
|
| 1263 |
+
"loss": 0.1168,
|
| 1264 |
+
"step": 3225
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 1.17,
|
| 1268 |
+
"grad_norm": 1.8701483011245728,
|
| 1269 |
+
"learning_rate": 8.287053732419763e-06,
|
| 1270 |
+
"loss": 0.1184,
|
| 1271 |
+
"step": 3250
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 1.18,
|
| 1275 |
+
"grad_norm": 1.9784027338027954,
|
| 1276 |
+
"learning_rate": 8.196898665705013e-06,
|
| 1277 |
+
"loss": 0.1085,
|
| 1278 |
+
"step": 3275
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 1.19,
|
| 1282 |
+
"grad_norm": 3.046462059020996,
|
| 1283 |
+
"learning_rate": 8.106743598990264e-06,
|
| 1284 |
+
"loss": 0.1282,
|
| 1285 |
+
"step": 3300
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 1.19,
|
| 1289 |
+
"eval_loss": 0.195108100771904,
|
| 1290 |
+
"eval_na_accuracy": 0.8539518713951111,
|
| 1291 |
+
"eval_ordinal_accuracy": 0.6971487402915955,
|
| 1292 |
+
"eval_ordinal_mae": 0.412724107503891,
|
| 1293 |
+
"eval_runtime": 175.5444,
|
| 1294 |
+
"eval_samples_per_second": 25.492,
|
| 1295 |
+
"eval_steps_per_second": 3.19,
|
| 1296 |
+
"step": 3300
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.2,
|
| 1300 |
+
"grad_norm": 1.4622838497161865,
|
| 1301 |
+
"learning_rate": 8.016588532275515e-06,
|
| 1302 |
+
"loss": 0.1428,
|
| 1303 |
+
"step": 3325
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 1.21,
|
| 1307 |
+
"grad_norm": 1.1039246320724487,
|
| 1308 |
+
"learning_rate": 7.926433465560766e-06,
|
| 1309 |
+
"loss": 0.1027,
|
| 1310 |
+
"step": 3350
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 1.22,
|
| 1314 |
+
"grad_norm": 1.006919503211975,
|
| 1315 |
+
"learning_rate": 7.836278398846015e-06,
|
| 1316 |
+
"loss": 0.1545,
|
| 1317 |
+
"step": 3375
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 1.23,
|
| 1321 |
+
"grad_norm": 0.9952992796897888,
|
| 1322 |
+
"learning_rate": 7.746123332131266e-06,
|
| 1323 |
+
"loss": 0.1406,
|
| 1324 |
+
"step": 3400
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 1.23,
|
| 1328 |
+
"eval_loss": 0.19589702785015106,
|
| 1329 |
+
"eval_na_accuracy": 0.8505154848098755,
|
| 1330 |
+
"eval_ordinal_accuracy": 0.6974055767059326,
|
| 1331 |
+
"eval_ordinal_mae": 0.4036160707473755,
|
| 1332 |
+
"eval_runtime": 176.1463,
|
| 1333 |
+
"eval_samples_per_second": 25.405,
|
| 1334 |
+
"eval_steps_per_second": 3.179,
|
| 1335 |
+
"step": 3400
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 1.24,
|
| 1339 |
+
"grad_norm": 2.206535577774048,
|
| 1340 |
+
"learning_rate": 7.655968265416517e-06,
|
| 1341 |
+
"loss": 0.1523,
|
| 1342 |
+
"step": 3425
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 1.24,
|
| 1346 |
+
"grad_norm": 2.8195924758911133,
|
| 1347 |
+
"learning_rate": 7.565813198701768e-06,
|
| 1348 |
+
"loss": 0.1357,
|
| 1349 |
+
"step": 3450
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.25,
|
| 1353 |
+
"grad_norm": 1.383612871170044,
|
| 1354 |
+
"learning_rate": 7.475658131987018e-06,
|
| 1355 |
+
"loss": 0.1262,
|
| 1356 |
+
"step": 3475
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.26,
|
| 1360 |
+
"grad_norm": 1.5167522430419922,
|
| 1361 |
+
"learning_rate": 7.385503065272269e-06,
|
| 1362 |
+
"loss": 0.0929,
|
| 1363 |
+
"step": 3500
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 1.26,
|
| 1367 |
+
"eval_loss": 0.19685061275959015,
|
| 1368 |
+
"eval_na_accuracy": 0.8453608155250549,
|
| 1369 |
+
"eval_ordinal_accuracy": 0.6976624727249146,
|
| 1370 |
+
"eval_ordinal_mae": 0.4020155370235443,
|
| 1371 |
+
"eval_runtime": 170.2776,
|
| 1372 |
+
"eval_samples_per_second": 26.281,
|
| 1373 |
+
"eval_steps_per_second": 3.289,
|
| 1374 |
+
"step": 3500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.27,
|
| 1378 |
+
"grad_norm": 0.9208484888076782,
|
| 1379 |
+
"learning_rate": 7.295347998557519e-06,
|
| 1380 |
+
"loss": 0.1193,
|
| 1381 |
+
"step": 3525
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.28,
|
| 1385 |
+
"grad_norm": 1.8087962865829468,
|
| 1386 |
+
"learning_rate": 7.205192931842771e-06,
|
| 1387 |
+
"loss": 0.1109,
|
| 1388 |
+
"step": 3550
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.29,
|
| 1392 |
+
"grad_norm": 1.2812789678573608,
|
| 1393 |
+
"learning_rate": 7.115037865128021e-06,
|
| 1394 |
+
"loss": 0.1778,
|
| 1395 |
+
"step": 3575
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.3,
|
| 1399 |
+
"grad_norm": 0.8966251015663147,
|
| 1400 |
+
"learning_rate": 7.0248827984132715e-06,
|
| 1401 |
+
"loss": 0.1135,
|
| 1402 |
+
"step": 3600
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.3,
|
| 1406 |
+
"eval_loss": 0.19572028517723083,
|
| 1407 |
+
"eval_na_accuracy": 0.831615149974823,
|
| 1408 |
+
"eval_ordinal_accuracy": 0.6981762051582336,
|
| 1409 |
+
"eval_ordinal_mae": 0.4026087522506714,
|
| 1410 |
+
"eval_runtime": 171.6945,
|
| 1411 |
+
"eval_samples_per_second": 26.064,
|
| 1412 |
+
"eval_steps_per_second": 3.262,
|
| 1413 |
+
"step": 3600
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 1.31,
|
| 1417 |
+
"grad_norm": 0.8011260628700256,
|
| 1418 |
+
"learning_rate": 6.9347277316985216e-06,
|
| 1419 |
+
"loss": 0.1078,
|
| 1420 |
+
"step": 3625
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 1.32,
|
| 1424 |
+
"grad_norm": 0.5675652623176575,
|
| 1425 |
+
"learning_rate": 6.844572664983773e-06,
|
| 1426 |
+
"loss": 0.1114,
|
| 1427 |
+
"step": 3650
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 1.33,
|
| 1431 |
+
"grad_norm": 1.0519170761108398,
|
| 1432 |
+
"learning_rate": 6.754417598269023e-06,
|
| 1433 |
+
"loss": 0.1435,
|
| 1434 |
+
"step": 3675
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 1.33,
|
| 1438 |
+
"grad_norm": 1.6201542615890503,
|
| 1439 |
+
"learning_rate": 6.664262531554274e-06,
|
| 1440 |
+
"loss": 0.1345,
|
| 1441 |
+
"step": 3700
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.33,
|
| 1445 |
+
"eval_loss": 0.1987382471561432,
|
| 1446 |
+
"eval_na_accuracy": 0.8814433217048645,
|
| 1447 |
+
"eval_ordinal_accuracy": 0.6832776665687561,
|
| 1448 |
+
"eval_ordinal_mae": 0.4107266366481781,
|
| 1449 |
+
"eval_runtime": 173.6501,
|
| 1450 |
+
"eval_samples_per_second": 25.77,
|
| 1451 |
+
"eval_steps_per_second": 3.225,
|
| 1452 |
+
"step": 3700
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.34,
|
| 1456 |
+
"grad_norm": 8.633369445800781,
|
| 1457 |
+
"learning_rate": 6.574107464839524e-06,
|
| 1458 |
+
"loss": 0.174,
|
| 1459 |
+
"step": 3725
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.35,
|
| 1463 |
+
"grad_norm": 1.3483631610870361,
|
| 1464 |
+
"learning_rate": 6.483952398124775e-06,
|
| 1465 |
+
"loss": 0.1104,
|
| 1466 |
+
"step": 3750
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.36,
|
| 1470 |
+
"grad_norm": 2.3553450107574463,
|
| 1471 |
+
"learning_rate": 6.393797331410025e-06,
|
| 1472 |
+
"loss": 0.1401,
|
| 1473 |
+
"step": 3775
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.37,
|
| 1477 |
+
"grad_norm": 6.273257732391357,
|
| 1478 |
+
"learning_rate": 6.303642264695277e-06,
|
| 1479 |
+
"loss": 0.1198,
|
| 1480 |
+
"step": 3800
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.37,
|
| 1484 |
+
"eval_loss": 0.19685617089271545,
|
| 1485 |
+
"eval_na_accuracy": 0.8522336483001709,
|
| 1486 |
+
"eval_ordinal_accuracy": 0.6992036700248718,
|
| 1487 |
+
"eval_ordinal_mae": 0.3987964391708374,
|
| 1488 |
+
"eval_runtime": 176.5744,
|
| 1489 |
+
"eval_samples_per_second": 25.343,
|
| 1490 |
+
"eval_steps_per_second": 3.171,
|
| 1491 |
+
"step": 3800
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 1.38,
|
| 1495 |
+
"grad_norm": 0.8371938467025757,
|
| 1496 |
+
"learning_rate": 6.213487197980527e-06,
|
| 1497 |
+
"loss": 0.1131,
|
| 1498 |
+
"step": 3825
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.39,
|
| 1502 |
+
"grad_norm": 4.081541061401367,
|
| 1503 |
+
"learning_rate": 6.123332131265778e-06,
|
| 1504 |
+
"loss": 0.1345,
|
| 1505 |
+
"step": 3850
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.4,
|
| 1509 |
+
"grad_norm": 2.648669958114624,
|
| 1510 |
+
"learning_rate": 6.033177064551028e-06,
|
| 1511 |
+
"loss": 0.1376,
|
| 1512 |
+
"step": 3875
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.41,
|
| 1516 |
+
"grad_norm": 0.8278887867927551,
|
| 1517 |
+
"learning_rate": 5.943021997836279e-06,
|
| 1518 |
+
"loss": 0.1281,
|
| 1519 |
+
"step": 3900
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 1.41,
|
| 1523 |
+
"eval_loss": 0.19765153527259827,
|
| 1524 |
+
"eval_na_accuracy": 0.8402062058448792,
|
| 1525 |
+
"eval_ordinal_accuracy": 0.6966350078582764,
|
| 1526 |
+
"eval_ordinal_mae": 0.4065835475921631,
|
| 1527 |
+
"eval_runtime": 174.9801,
|
| 1528 |
+
"eval_samples_per_second": 25.574,
|
| 1529 |
+
"eval_steps_per_second": 3.2,
|
| 1530 |
+
"step": 3900
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 1.42,
|
| 1534 |
+
"grad_norm": 0.9070908427238464,
|
| 1535 |
+
"learning_rate": 5.852866931121529e-06,
|
| 1536 |
+
"loss": 0.1278,
|
| 1537 |
+
"step": 3925
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 1.42,
|
| 1541 |
+
"grad_norm": 0.7849236130714417,
|
| 1542 |
+
"learning_rate": 5.7627118644067805e-06,
|
| 1543 |
+
"loss": 0.1575,
|
| 1544 |
+
"step": 3950
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 1.43,
|
| 1548 |
+
"grad_norm": 2.2364354133605957,
|
| 1549 |
+
"learning_rate": 5.6761630003606205e-06,
|
| 1550 |
+
"loss": 0.116,
|
| 1551 |
+
"step": 3975
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 1.44,
|
| 1555 |
+
"grad_norm": 0.8048083782196045,
|
| 1556 |
+
"learning_rate": 5.586007933645871e-06,
|
| 1557 |
+
"loss": 0.1153,
|
| 1558 |
+
"step": 4000
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 1.44,
|
| 1562 |
+
"eval_loss": 0.20141063630580902,
|
| 1563 |
+
"eval_na_accuracy": 0.8436425924301147,
|
| 1564 |
+
"eval_ordinal_accuracy": 0.693552553653717,
|
| 1565 |
+
"eval_ordinal_mae": 0.4091172218322754,
|
| 1566 |
+
"eval_runtime": 175.0141,
|
| 1567 |
+
"eval_samples_per_second": 25.569,
|
| 1568 |
+
"eval_steps_per_second": 3.2,
|
| 1569 |
+
"step": 4000
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 1.45,
|
| 1573 |
+
"grad_norm": 1.209535002708435,
|
| 1574 |
+
"learning_rate": 5.495852866931121e-06,
|
| 1575 |
+
"loss": 0.106,
|
| 1576 |
+
"step": 4025
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 1.46,
|
| 1580 |
+
"grad_norm": 3.71276593208313,
|
| 1581 |
+
"learning_rate": 5.405697800216373e-06,
|
| 1582 |
+
"loss": 0.1316,
|
| 1583 |
+
"step": 4050
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 1.47,
|
| 1587 |
+
"grad_norm": 18.610342025756836,
|
| 1588 |
+
"learning_rate": 5.315542733501623e-06,
|
| 1589 |
+
"loss": 0.132,
|
| 1590 |
+
"step": 4075
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 1.48,
|
| 1594 |
+
"grad_norm": 2.119901180267334,
|
| 1595 |
+
"learning_rate": 5.225387666786874e-06,
|
| 1596 |
+
"loss": 0.1485,
|
| 1597 |
+
"step": 4100
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 1.48,
|
| 1601 |
+
"eval_loss": 0.19648495316505432,
|
| 1602 |
+
"eval_na_accuracy": 0.838487982749939,
|
| 1603 |
+
"eval_ordinal_accuracy": 0.7038273811340332,
|
| 1604 |
+
"eval_ordinal_mae": 0.39893215894699097,
|
| 1605 |
+
"eval_runtime": 174.9887,
|
| 1606 |
+
"eval_samples_per_second": 25.573,
|
| 1607 |
+
"eval_steps_per_second": 3.2,
|
| 1608 |
+
"step": 4100
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 1.49,
|
| 1612 |
+
"grad_norm": 1.489436388015747,
|
| 1613 |
+
"learning_rate": 5.135232600072124e-06,
|
| 1614 |
+
"loss": 0.1158,
|
| 1615 |
+
"step": 4125
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 1.5,
|
| 1619 |
+
"grad_norm": 3.2724928855895996,
|
| 1620 |
+
"learning_rate": 5.045077533357375e-06,
|
| 1621 |
+
"loss": 0.1098,
|
| 1622 |
+
"step": 4150
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 1.51,
|
| 1626 |
+
"grad_norm": 1.0074132680892944,
|
| 1627 |
+
"learning_rate": 4.954922466642626e-06,
|
| 1628 |
+
"loss": 0.1094,
|
| 1629 |
+
"step": 4175
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 1.51,
|
| 1633 |
+
"grad_norm": 0.7578648924827576,
|
| 1634 |
+
"learning_rate": 4.864767399927877e-06,
|
| 1635 |
+
"loss": 0.1292,
|
| 1636 |
+
"step": 4200
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 1.51,
|
| 1640 |
+
"eval_loss": 0.19693247973918915,
|
| 1641 |
+
"eval_na_accuracy": 0.8470790386199951,
|
| 1642 |
+
"eval_ordinal_accuracy": 0.7030567526817322,
|
| 1643 |
+
"eval_ordinal_mae": 0.39781150221824646,
|
| 1644 |
+
"eval_runtime": 173.0859,
|
| 1645 |
+
"eval_samples_per_second": 25.854,
|
| 1646 |
+
"eval_steps_per_second": 3.235,
|
| 1647 |
+
"step": 4200
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 1.52,
|
| 1651 |
+
"grad_norm": 3.7381861209869385,
|
| 1652 |
+
"learning_rate": 4.774612333213128e-06,
|
| 1653 |
+
"loss": 0.1309,
|
| 1654 |
+
"step": 4225
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 1.53,
|
| 1658 |
+
"grad_norm": 0.8151502013206482,
|
| 1659 |
+
"learning_rate": 4.684457266498378e-06,
|
| 1660 |
+
"loss": 0.1428,
|
| 1661 |
+
"step": 4250
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 1.54,
|
| 1665 |
+
"grad_norm": 0.9021445512771606,
|
| 1666 |
+
"learning_rate": 4.5943021997836285e-06,
|
| 1667 |
+
"loss": 0.1104,
|
| 1668 |
+
"step": 4275
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 1.55,
|
| 1672 |
+
"grad_norm": 3.8359127044677734,
|
| 1673 |
+
"learning_rate": 4.5041471330688794e-06,
|
| 1674 |
+
"loss": 0.1233,
|
| 1675 |
+
"step": 4300
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 1.55,
|
| 1679 |
+
"eval_loss": 0.1989138424396515,
|
| 1680 |
+
"eval_na_accuracy": 0.8659793734550476,
|
| 1681 |
+
"eval_ordinal_accuracy": 0.6950937509536743,
|
| 1682 |
+
"eval_ordinal_mae": 0.39926186203956604,
|
| 1683 |
+
"eval_runtime": 173.9,
|
| 1684 |
+
"eval_samples_per_second": 25.733,
|
| 1685 |
+
"eval_steps_per_second": 3.22,
|
| 1686 |
+
"step": 4300
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 1.56,
|
| 1690 |
+
"grad_norm": 2.0737252235412598,
|
| 1691 |
+
"learning_rate": 4.4139920663541295e-06,
|
| 1692 |
+
"loss": 0.1343,
|
| 1693 |
+
"step": 4325
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 1.57,
|
| 1697 |
+
"grad_norm": 0.8500555753707886,
|
| 1698 |
+
"learning_rate": 4.32383699963938e-06,
|
| 1699 |
+
"loss": 0.1274,
|
| 1700 |
+
"step": 4350
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 1.58,
|
| 1704 |
+
"grad_norm": 4.984882354736328,
|
| 1705 |
+
"learning_rate": 4.233681932924631e-06,
|
| 1706 |
+
"loss": 0.1131,
|
| 1707 |
+
"step": 4375
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 1.59,
|
| 1711 |
+
"grad_norm": 0.8260248899459839,
|
| 1712 |
+
"learning_rate": 4.143526866209881e-06,
|
| 1713 |
+
"loss": 0.1128,
|
| 1714 |
+
"step": 4400
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 1.59,
|
| 1718 |
+
"eval_loss": 0.1998443454504013,
|
| 1719 |
+
"eval_na_accuracy": 0.8522336483001709,
|
| 1720 |
+
"eval_ordinal_accuracy": 0.6971487402915955,
|
| 1721 |
+
"eval_ordinal_mae": 0.391990065574646,
|
| 1722 |
+
"eval_runtime": 175.0388,
|
| 1723 |
+
"eval_samples_per_second": 25.566,
|
| 1724 |
+
"eval_steps_per_second": 3.199,
|
| 1725 |
+
"step": 4400
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.6,
|
| 1729 |
+
"grad_norm": 0.7347471117973328,
|
| 1730 |
+
"learning_rate": 4.053371799495132e-06,
|
| 1731 |
+
"loss": 0.1095,
|
| 1732 |
+
"step": 4425
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.6,
|
| 1736 |
+
"grad_norm": 0.6527581810951233,
|
| 1737 |
+
"learning_rate": 3.963216732780383e-06,
|
| 1738 |
+
"loss": 0.1627,
|
| 1739 |
+
"step": 4450
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.61,
|
| 1743 |
+
"grad_norm": 1.0017273426055908,
|
| 1744 |
+
"learning_rate": 3.873061666065633e-06,
|
| 1745 |
+
"loss": 0.1047,
|
| 1746 |
+
"step": 4475
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.62,
|
| 1750 |
+
"grad_norm": 1.7798997163772583,
|
| 1751 |
+
"learning_rate": 3.782906599350884e-06,
|
| 1752 |
+
"loss": 0.0964,
|
| 1753 |
+
"step": 4500
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.62,
|
| 1757 |
+
"eval_loss": 0.20046481490135193,
|
| 1758 |
+
"eval_na_accuracy": 0.8625429272651672,
|
| 1759 |
+
"eval_ordinal_accuracy": 0.6981762051582336,
|
| 1760 |
+
"eval_ordinal_mae": 0.39257487654685974,
|
| 1761 |
+
"eval_runtime": 174.4806,
|
| 1762 |
+
"eval_samples_per_second": 25.648,
|
| 1763 |
+
"eval_steps_per_second": 3.21,
|
| 1764 |
+
"step": 4500
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 1.63,
|
| 1768 |
+
"grad_norm": 0.8608025312423706,
|
| 1769 |
+
"learning_rate": 3.6927515326361344e-06,
|
| 1770 |
+
"loss": 0.0974,
|
| 1771 |
+
"step": 4525
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 1.64,
|
| 1775 |
+
"grad_norm": 0.5464763641357422,
|
| 1776 |
+
"learning_rate": 3.6025964659213853e-06,
|
| 1777 |
+
"loss": 0.1224,
|
| 1778 |
+
"step": 4550
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 1.65,
|
| 1782 |
+
"grad_norm": 1.151129961013794,
|
| 1783 |
+
"learning_rate": 3.5124413992066358e-06,
|
| 1784 |
+
"loss": 0.1514,
|
| 1785 |
+
"step": 4575
|
| 1786 |
+
},
|
| 1787 |
+
{
|
| 1788 |
+
"epoch": 1.66,
|
| 1789 |
+
"grad_norm": 0.5751986503601074,
|
| 1790 |
+
"learning_rate": 3.4222863324918867e-06,
|
| 1791 |
+
"loss": 0.1184,
|
| 1792 |
+
"step": 4600
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 1.66,
|
| 1796 |
+
"eval_loss": 0.20076720416545868,
|
| 1797 |
+
"eval_na_accuracy": 0.8711340427398682,
|
| 1798 |
+
"eval_ordinal_accuracy": 0.6968918442726135,
|
| 1799 |
+
"eval_ordinal_mae": 0.3860225975513458,
|
| 1800 |
+
"eval_runtime": 175.6222,
|
| 1801 |
+
"eval_samples_per_second": 25.481,
|
| 1802 |
+
"eval_steps_per_second": 3.189,
|
| 1803 |
+
"step": 4600
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 1.67,
|
| 1807 |
+
"grad_norm": 0.8153491020202637,
|
| 1808 |
+
"learning_rate": 3.332131265777137e-06,
|
| 1809 |
+
"loss": 0.1344,
|
| 1810 |
+
"step": 4625
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 1.68,
|
| 1814 |
+
"grad_norm": 1.3030270338058472,
|
| 1815 |
+
"learning_rate": 3.2419761990623876e-06,
|
| 1816 |
+
"loss": 0.136,
|
| 1817 |
+
"step": 4650
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 1.69,
|
| 1821 |
+
"grad_norm": 6.30877685546875,
|
| 1822 |
+
"learning_rate": 3.1518211323476385e-06,
|
| 1823 |
+
"loss": 0.1224,
|
| 1824 |
+
"step": 4675
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 1.69,
|
| 1828 |
+
"grad_norm": 4.972422122955322,
|
| 1829 |
+
"learning_rate": 3.061666065632889e-06,
|
| 1830 |
+
"loss": 0.108,
|
| 1831 |
+
"step": 4700
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 1.69,
|
| 1835 |
+
"eval_loss": 0.19936376810073853,
|
| 1836 |
+
"eval_na_accuracy": 0.8573883175849915,
|
| 1837 |
+
"eval_ordinal_accuracy": 0.702029287815094,
|
| 1838 |
+
"eval_ordinal_mae": 0.39070162177085876,
|
| 1839 |
+
"eval_runtime": 174.219,
|
| 1840 |
+
"eval_samples_per_second": 25.686,
|
| 1841 |
+
"eval_steps_per_second": 3.214,
|
| 1842 |
+
"step": 4700
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 1.7,
|
| 1846 |
+
"grad_norm": 1.4122976064682007,
|
| 1847 |
+
"learning_rate": 2.9715109989181394e-06,
|
| 1848 |
+
"loss": 0.1496,
|
| 1849 |
+
"step": 4725
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 1.71,
|
| 1853 |
+
"grad_norm": 2.4173316955566406,
|
| 1854 |
+
"learning_rate": 2.8813559322033903e-06,
|
| 1855 |
+
"loss": 0.1276,
|
| 1856 |
+
"step": 4750
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 1.72,
|
| 1860 |
+
"grad_norm": 1.0301971435546875,
|
| 1861 |
+
"learning_rate": 2.7912008654886407e-06,
|
| 1862 |
+
"loss": 0.1315,
|
| 1863 |
+
"step": 4775
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 1.73,
|
| 1867 |
+
"grad_norm": 0.8369082808494568,
|
| 1868 |
+
"learning_rate": 2.701045798773891e-06,
|
| 1869 |
+
"loss": 0.129,
|
| 1870 |
+
"step": 4800
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 1.73,
|
| 1874 |
+
"eval_loss": 0.19854028522968292,
|
| 1875 |
+
"eval_na_accuracy": 0.8591065406799316,
|
| 1876 |
+
"eval_ordinal_accuracy": 0.7033136487007141,
|
| 1877 |
+
"eval_ordinal_mae": 0.3896414041519165,
|
| 1878 |
+
"eval_runtime": 176.1539,
|
| 1879 |
+
"eval_samples_per_second": 25.404,
|
| 1880 |
+
"eval_steps_per_second": 3.179,
|
| 1881 |
+
"step": 4800
|
| 1882 |
+
},
|
| 1883 |
+
{
|
| 1884 |
+
"epoch": 1.74,
|
| 1885 |
+
"grad_norm": 3.947204113006592,
|
| 1886 |
+
"learning_rate": 2.610890732059142e-06,
|
| 1887 |
+
"loss": 0.1106,
|
| 1888 |
+
"step": 4825
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 1.75,
|
| 1892 |
+
"grad_norm": 2.09318470954895,
|
| 1893 |
+
"learning_rate": 2.5207356653443925e-06,
|
| 1894 |
+
"loss": 0.164,
|
| 1895 |
+
"step": 4850
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 1.76,
|
| 1899 |
+
"grad_norm": 0.7661357522010803,
|
| 1900 |
+
"learning_rate": 2.4305805986296434e-06,
|
| 1901 |
+
"loss": 0.1214,
|
| 1902 |
+
"step": 4875
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 1.77,
|
| 1906 |
+
"grad_norm": 0.4803534746170044,
|
| 1907 |
+
"learning_rate": 2.340425531914894e-06,
|
| 1908 |
+
"loss": 0.1396,
|
| 1909 |
+
"step": 4900
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 1.77,
|
| 1913 |
+
"eval_loss": 0.19975444674491882,
|
| 1914 |
+
"eval_na_accuracy": 0.8573883175849915,
|
| 1915 |
+
"eval_ordinal_accuracy": 0.6984331011772156,
|
| 1916 |
+
"eval_ordinal_mae": 0.3833589553833008,
|
| 1917 |
+
"eval_runtime": 176.3204,
|
| 1918 |
+
"eval_samples_per_second": 25.38,
|
| 1919 |
+
"eval_steps_per_second": 3.176,
|
| 1920 |
+
"step": 4900
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 1.78,
|
| 1924 |
+
"grad_norm": 0.7548712491989136,
|
| 1925 |
+
"learning_rate": 2.2502704652001443e-06,
|
| 1926 |
+
"loss": 0.0977,
|
| 1927 |
+
"step": 4925
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 1.79,
|
| 1931 |
+
"grad_norm": 1.906633734703064,
|
| 1932 |
+
"learning_rate": 2.1601153984853952e-06,
|
| 1933 |
+
"loss": 0.1385,
|
| 1934 |
+
"step": 4950
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 1.79,
|
| 1938 |
+
"grad_norm": 0.6955749988555908,
|
| 1939 |
+
"learning_rate": 2.0699603317706457e-06,
|
| 1940 |
+
"loss": 0.1017,
|
| 1941 |
+
"step": 4975
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 1.8,
|
| 1945 |
+
"grad_norm": 6.211479663848877,
|
| 1946 |
+
"learning_rate": 1.979805265055896e-06,
|
| 1947 |
+
"loss": 0.1323,
|
| 1948 |
+
"step": 5000
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 1.8,
|
| 1952 |
+
"eval_loss": 0.19856412708759308,
|
| 1953 |
+
"eval_na_accuracy": 0.8453608155250549,
|
| 1954 |
+
"eval_ordinal_accuracy": 0.7051117420196533,
|
| 1955 |
+
"eval_ordinal_mae": 0.3843817710876465,
|
| 1956 |
+
"eval_runtime": 173.9748,
|
| 1957 |
+
"eval_samples_per_second": 25.722,
|
| 1958 |
+
"eval_steps_per_second": 3.219,
|
| 1959 |
+
"step": 5000
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 1.81,
|
| 1963 |
+
"grad_norm": 0.5422704219818115,
|
| 1964 |
+
"learning_rate": 1.889650198341147e-06,
|
| 1965 |
+
"loss": 0.1367,
|
| 1966 |
+
"step": 5025
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 1.82,
|
| 1970 |
+
"grad_norm": 2.782160997390747,
|
| 1971 |
+
"learning_rate": 1.7994951316263975e-06,
|
| 1972 |
+
"loss": 0.1002,
|
| 1973 |
+
"step": 5050
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 1.83,
|
| 1977 |
+
"grad_norm": 4.336767673492432,
|
| 1978 |
+
"learning_rate": 1.7093400649116482e-06,
|
| 1979 |
+
"loss": 0.1407,
|
| 1980 |
+
"step": 5075
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 1.84,
|
| 1984 |
+
"grad_norm": 3.1643247604370117,
|
| 1985 |
+
"learning_rate": 1.6191849981968988e-06,
|
| 1986 |
+
"loss": 0.1079,
|
| 1987 |
+
"step": 5100
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 1.84,
|
| 1991 |
+
"eval_loss": 0.19739079475402832,
|
| 1992 |
+
"eval_na_accuracy": 0.8402062058448792,
|
| 1993 |
+
"eval_ordinal_accuracy": 0.7053686380386353,
|
| 1994 |
+
"eval_ordinal_mae": 0.3832775950431824,
|
| 1995 |
+
"eval_runtime": 174.8648,
|
| 1996 |
+
"eval_samples_per_second": 25.591,
|
| 1997 |
+
"eval_steps_per_second": 3.202,
|
| 1998 |
+
"step": 5100
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.85,
|
| 2002 |
+
"grad_norm": 5.217372894287109,
|
| 2003 |
+
"learning_rate": 1.5290299314821493e-06,
|
| 2004 |
+
"loss": 0.122,
|
| 2005 |
+
"step": 5125
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.86,
|
| 2009 |
+
"grad_norm": 4.001213073730469,
|
| 2010 |
+
"learning_rate": 1.4388748647674e-06,
|
| 2011 |
+
"loss": 0.1102,
|
| 2012 |
+
"step": 5150
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.87,
|
| 2016 |
+
"grad_norm": 2.418731451034546,
|
| 2017 |
+
"learning_rate": 1.3487197980526506e-06,
|
| 2018 |
+
"loss": 0.1117,
|
| 2019 |
+
"step": 5175
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.88,
|
| 2023 |
+
"grad_norm": 1.181628704071045,
|
| 2024 |
+
"learning_rate": 1.2585647313379013e-06,
|
| 2025 |
+
"loss": 0.0802,
|
| 2026 |
+
"step": 5200
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.88,
|
| 2030 |
+
"eval_loss": 0.19647686183452606,
|
| 2031 |
+
"eval_na_accuracy": 0.8487972617149353,
|
| 2032 |
+
"eval_ordinal_accuracy": 0.7074235677719116,
|
| 2033 |
+
"eval_ordinal_mae": 0.38217073678970337,
|
| 2034 |
+
"eval_runtime": 175.4662,
|
| 2035 |
+
"eval_samples_per_second": 25.503,
|
| 2036 |
+
"eval_steps_per_second": 3.191,
|
| 2037 |
+
"step": 5200
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"epoch": 1.88,
|
| 2041 |
+
"grad_norm": 0.654973030090332,
|
| 2042 |
+
"learning_rate": 1.1684096646231518e-06,
|
| 2043 |
+
"loss": 0.1139,
|
| 2044 |
+
"step": 5225
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 1.89,
|
| 2048 |
+
"grad_norm": 0.66736900806427,
|
| 2049 |
+
"learning_rate": 1.0782545979084025e-06,
|
| 2050 |
+
"loss": 0.1099,
|
| 2051 |
+
"step": 5250
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 1.9,
|
| 2055 |
+
"grad_norm": 0.9291324615478516,
|
| 2056 |
+
"learning_rate": 9.880995311936533e-07,
|
| 2057 |
+
"loss": 0.1233,
|
| 2058 |
+
"step": 5275
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 1.91,
|
| 2062 |
+
"grad_norm": 0.864493191242218,
|
| 2063 |
+
"learning_rate": 8.979444644789038e-07,
|
| 2064 |
+
"loss": 0.1391,
|
| 2065 |
+
"step": 5300
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 1.91,
|
| 2069 |
+
"eval_loss": 0.19748477637767792,
|
| 2070 |
+
"eval_na_accuracy": 0.8453608155250549,
|
| 2071 |
+
"eval_ordinal_accuracy": 0.7051117420196533,
|
| 2072 |
+
"eval_ordinal_mae": 0.3809069097042084,
|
| 2073 |
+
"eval_runtime": 174.5451,
|
| 2074 |
+
"eval_samples_per_second": 25.638,
|
| 2075 |
+
"eval_steps_per_second": 3.208,
|
| 2076 |
+
"step": 5300
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 1.92,
|
| 2080 |
+
"grad_norm": 0.844887912273407,
|
| 2081 |
+
"learning_rate": 8.077893977641545e-07,
|
| 2082 |
+
"loss": 0.1364,
|
| 2083 |
+
"step": 5325
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 1.93,
|
| 2087 |
+
"grad_norm": 0.7487301230430603,
|
| 2088 |
+
"learning_rate": 7.17634331049405e-07,
|
| 2089 |
+
"loss": 0.1137,
|
| 2090 |
+
"step": 5350
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.94,
|
| 2094 |
+
"grad_norm": 3.6033549308776855,
|
| 2095 |
+
"learning_rate": 6.274792643346557e-07,
|
| 2096 |
+
"loss": 0.1508,
|
| 2097 |
+
"step": 5375
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 1.95,
|
| 2101 |
+
"grad_norm": 0.7312902212142944,
|
| 2102 |
+
"learning_rate": 5.373241976199063e-07,
|
| 2103 |
+
"loss": 0.1183,
|
| 2104 |
+
"step": 5400
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 1.95,
|
| 2108 |
+
"eval_loss": 0.19734890758991241,
|
| 2109 |
+
"eval_na_accuracy": 0.8350515365600586,
|
| 2110 |
+
"eval_ordinal_accuracy": 0.7087079286575317,
|
| 2111 |
+
"eval_ordinal_mae": 0.38272929191589355,
|
| 2112 |
+
"eval_runtime": 176.207,
|
| 2113 |
+
"eval_samples_per_second": 25.396,
|
| 2114 |
+
"eval_steps_per_second": 3.178,
|
| 2115 |
+
"step": 5400
|
| 2116 |
+
},
|
| 2117 |
+
{
|
| 2118 |
+
"epoch": 1.96,
|
| 2119 |
+
"grad_norm": 0.859944760799408,
|
| 2120 |
+
"learning_rate": 4.4716913090515695e-07,
|
| 2121 |
+
"loss": 0.1125,
|
| 2122 |
+
"step": 5425
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 1.97,
|
| 2126 |
+
"grad_norm": 0.7725034952163696,
|
| 2127 |
+
"learning_rate": 3.5701406419040757e-07,
|
| 2128 |
+
"loss": 0.1009,
|
| 2129 |
+
"step": 5450
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 1.97,
|
| 2133 |
+
"grad_norm": 8.944540977478027,
|
| 2134 |
+
"learning_rate": 2.6685899747565814e-07,
|
| 2135 |
+
"loss": 0.1115,
|
| 2136 |
+
"step": 5475
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 1.98,
|
| 2140 |
+
"grad_norm": 0.6212058663368225,
|
| 2141 |
+
"learning_rate": 1.7670393076090878e-07,
|
| 2142 |
+
"loss": 0.1368,
|
| 2143 |
+
"step": 5500
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 1.98,
|
| 2147 |
+
"eval_loss": 0.19747541844844818,
|
| 2148 |
+
"eval_na_accuracy": 0.8333333134651184,
|
| 2149 |
+
"eval_ordinal_accuracy": 0.7081941962242126,
|
| 2150 |
+
"eval_ordinal_mae": 0.38125094771385193,
|
| 2151 |
+
"eval_runtime": 170.5638,
|
| 2152 |
+
"eval_samples_per_second": 26.237,
|
| 2153 |
+
"eval_steps_per_second": 3.283,
|
| 2154 |
+
"step": 5500
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 1.99,
|
| 2158 |
+
"grad_norm": 0.4366958737373352,
|
| 2159 |
+
"learning_rate": 8.65488640461594e-08,
|
| 2160 |
+
"loss": 0.108,
|
| 2161 |
+
"step": 5525
|
| 2162 |
+
},
|
| 2163 |
+
{
|
| 2164 |
+
"epoch": 2.0,
|
| 2165 |
+
"step": 5546,
|
| 2166 |
+
"total_flos": 6.876575711894569e+18,
|
| 2167 |
+
"train_loss": 0.18857605492557122,
|
| 2168 |
+
"train_runtime": 17933.6413,
|
| 2169 |
+
"train_samples_per_second": 4.948,
|
| 2170 |
+
"train_steps_per_second": 0.309
|
| 2171 |
+
}
|
| 2172 |
+
],
|
| 2173 |
+
"logging_steps": 25,
|
| 2174 |
+
"max_steps": 5546,
|
| 2175 |
+
"num_input_tokens_seen": 0,
|
| 2176 |
+
"num_train_epochs": 2,
|
| 2177 |
+
"save_steps": 100,
|
| 2178 |
+
"total_flos": 6.876575711894569e+18,
|
| 2179 |
+
"train_batch_size": 16,
|
| 2180 |
+
"trial_name": null,
|
| 2181 |
+
"trial_params": null
|
| 2182 |
+
}
|