End of training
Browse files- all_results.json +15 -0
- eval_results.json +10 -0
- logs/events.out.tfevents.1674841090.serv-3328.2273911.2 +3 -0
- train_results.json +8 -0
- trainer_state.json +2545 -0
all_results.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 167.88,
|
3 |
+
"eval_accuracy": 0.46066022044935584,
|
4 |
+
"eval_loss": 2.9848647117614746,
|
5 |
+
"eval_runtime": 3.0046,
|
6 |
+
"eval_samples": 479,
|
7 |
+
"eval_samples_per_second": 159.423,
|
8 |
+
"eval_steps_per_second": 1.331,
|
9 |
+
"perplexity": 19.783825536403292,
|
10 |
+
"train_loss": 3.1306004736328124,
|
11 |
+
"train_runtime": 219827.3924,
|
12 |
+
"train_samples": 228639,
|
13 |
+
"train_samples_per_second": 174.683,
|
14 |
+
"train_steps_per_second": 1.365
|
15 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 167.88,
|
3 |
+
"eval_accuracy": 0.46066022044935584,
|
4 |
+
"eval_loss": 2.9848647117614746,
|
5 |
+
"eval_runtime": 3.0046,
|
6 |
+
"eval_samples": 479,
|
7 |
+
"eval_samples_per_second": 159.423,
|
8 |
+
"eval_steps_per_second": 1.331,
|
9 |
+
"perplexity": 19.783825536403292
|
10 |
+
}
|
logs/events.out.tfevents.1674841090.serv-3328.2273911.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f50a72775b9edc4cc67dfd03037c53ade87f196e411a0bafc9bf0abb6cc7dc02
|
3 |
+
size 369
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 167.88,
|
3 |
+
"train_loss": 3.1306004736328124,
|
4 |
+
"train_runtime": 219827.3924,
|
5 |
+
"train_samples": 228639,
|
6 |
+
"train_samples_per_second": 174.683,
|
7 |
+
"train_steps_per_second": 1.365
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2545 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 2.9971697330474854,
|
3 |
+
"best_model_checkpoint": "mobilebert_add_pre-training-complete/checkpoint-293068",
|
4 |
+
"epoch": 167.87912702853944,
|
5 |
+
"global_step": 300000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 1.0,
|
12 |
+
"learning_rate": 4.971873957985996e-05,
|
13 |
+
"loss": 4.8119,
|
14 |
+
"step": 1787
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 1.0,
|
18 |
+
"eval_accuracy": 0.31985974754558205,
|
19 |
+
"eval_loss": 4.369968414306641,
|
20 |
+
"eval_runtime": 3.3492,
|
21 |
+
"eval_samples_per_second": 143.018,
|
22 |
+
"eval_steps_per_second": 1.194,
|
23 |
+
"step": 1787
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 2.0,
|
27 |
+
"learning_rate": 4.942080693564522e-05,
|
28 |
+
"loss": 4.2649,
|
29 |
+
"step": 3574
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 2.0,
|
33 |
+
"eval_accuracy": 0.3444790701591847,
|
34 |
+
"eval_loss": 4.09299373626709,
|
35 |
+
"eval_runtime": 2.6635,
|
36 |
+
"eval_samples_per_second": 179.842,
|
37 |
+
"eval_steps_per_second": 1.502,
|
38 |
+
"step": 3574
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 3.0,
|
42 |
+
"learning_rate": 4.912287429143048e-05,
|
43 |
+
"loss": 4.0457,
|
44 |
+
"step": 5361
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 3.0,
|
48 |
+
"eval_accuracy": 0.3544818746691187,
|
49 |
+
"eval_loss": 3.937490224838257,
|
50 |
+
"eval_runtime": 2.6837,
|
51 |
+
"eval_samples_per_second": 178.487,
|
52 |
+
"eval_steps_per_second": 1.491,
|
53 |
+
"step": 5361
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 4.0,
|
57 |
+
"learning_rate": 4.882494164721574e-05,
|
58 |
+
"loss": 3.9099,
|
59 |
+
"step": 7148
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 4.0,
|
63 |
+
"eval_accuracy": 0.3643615542831731,
|
64 |
+
"eval_loss": 3.853400468826294,
|
65 |
+
"eval_runtime": 2.6447,
|
66 |
+
"eval_samples_per_second": 181.115,
|
67 |
+
"eval_steps_per_second": 1.512,
|
68 |
+
"step": 7148
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 5.0,
|
72 |
+
"learning_rate": 4.8527009003001e-05,
|
73 |
+
"loss": 3.8193,
|
74 |
+
"step": 8935
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 5.0,
|
78 |
+
"eval_accuracy": 0.3669499105545617,
|
79 |
+
"eval_loss": 3.7993216514587402,
|
80 |
+
"eval_runtime": 2.8341,
|
81 |
+
"eval_samples_per_second": 169.013,
|
82 |
+
"eval_steps_per_second": 1.411,
|
83 |
+
"step": 8935
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 6.0,
|
87 |
+
"learning_rate": 4.822907635878626e-05,
|
88 |
+
"loss": 3.7517,
|
89 |
+
"step": 10722
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 6.0,
|
93 |
+
"eval_accuracy": 0.3729702583561185,
|
94 |
+
"eval_loss": 3.7414000034332275,
|
95 |
+
"eval_runtime": 2.6482,
|
96 |
+
"eval_samples_per_second": 180.878,
|
97 |
+
"eval_steps_per_second": 1.51,
|
98 |
+
"step": 10722
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 7.0,
|
102 |
+
"learning_rate": 4.7931143714571526e-05,
|
103 |
+
"loss": 3.6983,
|
104 |
+
"step": 12509
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 7.0,
|
108 |
+
"eval_accuracy": 0.38179712437501767,
|
109 |
+
"eval_loss": 3.673661708831787,
|
110 |
+
"eval_runtime": 2.6433,
|
111 |
+
"eval_samples_per_second": 181.216,
|
112 |
+
"eval_steps_per_second": 1.513,
|
113 |
+
"step": 12509
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 8.0,
|
117 |
+
"learning_rate": 4.7633211070356786e-05,
|
118 |
+
"loss": 3.6565,
|
119 |
+
"step": 14296
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 8.0,
|
123 |
+
"eval_accuracy": 0.3794307568661142,
|
124 |
+
"eval_loss": 3.6657235622406006,
|
125 |
+
"eval_runtime": 2.6602,
|
126 |
+
"eval_samples_per_second": 180.063,
|
127 |
+
"eval_steps_per_second": 1.504,
|
128 |
+
"step": 14296
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 9.0,
|
132 |
+
"learning_rate": 4.7335278426142046e-05,
|
133 |
+
"loss": 3.619,
|
134 |
+
"step": 16083
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 9.0,
|
138 |
+
"eval_accuracy": 0.38690059135113924,
|
139 |
+
"eval_loss": 3.6128811836242676,
|
140 |
+
"eval_runtime": 2.9694,
|
141 |
+
"eval_samples_per_second": 161.314,
|
142 |
+
"eval_steps_per_second": 1.347,
|
143 |
+
"step": 16083
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 10.0,
|
147 |
+
"learning_rate": 4.703734578192731e-05,
|
148 |
+
"loss": 3.5899,
|
149 |
+
"step": 17870
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 10.0,
|
153 |
+
"eval_accuracy": 0.39099787685774945,
|
154 |
+
"eval_loss": 3.580448627471924,
|
155 |
+
"eval_runtime": 2.6649,
|
156 |
+
"eval_samples_per_second": 179.741,
|
157 |
+
"eval_steps_per_second": 1.501,
|
158 |
+
"step": 17870
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 11.0,
|
162 |
+
"learning_rate": 4.673941313771257e-05,
|
163 |
+
"loss": 3.5597,
|
164 |
+
"step": 19657
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 11.0,
|
168 |
+
"eval_accuracy": 0.3964062763624093,
|
169 |
+
"eval_loss": 3.5432231426239014,
|
170 |
+
"eval_runtime": 2.6576,
|
171 |
+
"eval_samples_per_second": 180.239,
|
172 |
+
"eval_steps_per_second": 1.505,
|
173 |
+
"step": 19657
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 12.0,
|
177 |
+
"learning_rate": 4.644148049349783e-05,
|
178 |
+
"loss": 3.5329,
|
179 |
+
"step": 21444
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 12.0,
|
183 |
+
"eval_accuracy": 0.39582186502256966,
|
184 |
+
"eval_loss": 3.539653778076172,
|
185 |
+
"eval_runtime": 2.6515,
|
186 |
+
"eval_samples_per_second": 180.65,
|
187 |
+
"eval_steps_per_second": 1.509,
|
188 |
+
"step": 21444
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 13.0,
|
192 |
+
"learning_rate": 4.61435478492831e-05,
|
193 |
+
"loss": 3.5088,
|
194 |
+
"step": 23231
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 13.0,
|
198 |
+
"eval_accuracy": 0.4010702757800555,
|
199 |
+
"eval_loss": 3.48960542678833,
|
200 |
+
"eval_runtime": 2.671,
|
201 |
+
"eval_samples_per_second": 179.334,
|
202 |
+
"eval_steps_per_second": 1.498,
|
203 |
+
"step": 23231
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 14.0,
|
207 |
+
"learning_rate": 4.584561520506836e-05,
|
208 |
+
"loss": 3.4904,
|
209 |
+
"step": 25018
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 14.0,
|
213 |
+
"eval_accuracy": 0.39997756086615055,
|
214 |
+
"eval_loss": 3.473088502883911,
|
215 |
+
"eval_runtime": 2.9516,
|
216 |
+
"eval_samples_per_second": 162.282,
|
217 |
+
"eval_steps_per_second": 1.355,
|
218 |
+
"step": 25018
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 15.0,
|
222 |
+
"learning_rate": 4.5547682560853625e-05,
|
223 |
+
"loss": 3.4703,
|
224 |
+
"step": 26805
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 15.0,
|
228 |
+
"eval_accuracy": 0.39938999029530015,
|
229 |
+
"eval_loss": 3.4971117973327637,
|
230 |
+
"eval_runtime": 2.873,
|
231 |
+
"eval_samples_per_second": 166.723,
|
232 |
+
"eval_steps_per_second": 1.392,
|
233 |
+
"step": 26805
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 16.0,
|
237 |
+
"learning_rate": 4.5249749916638885e-05,
|
238 |
+
"loss": 3.4533,
|
239 |
+
"step": 28592
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 16.0,
|
243 |
+
"eval_accuracy": 0.4049408937204818,
|
244 |
+
"eval_loss": 3.460916519165039,
|
245 |
+
"eval_runtime": 3.1252,
|
246 |
+
"eval_samples_per_second": 153.27,
|
247 |
+
"eval_steps_per_second": 1.28,
|
248 |
+
"step": 28592
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 17.0,
|
252 |
+
"learning_rate": 4.4951817272424145e-05,
|
253 |
+
"loss": 3.4369,
|
254 |
+
"step": 30379
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 17.0,
|
258 |
+
"eval_accuracy": 0.4066726276964346,
|
259 |
+
"eval_loss": 3.441118001937866,
|
260 |
+
"eval_runtime": 2.8852,
|
261 |
+
"eval_samples_per_second": 166.017,
|
262 |
+
"eval_steps_per_second": 1.386,
|
263 |
+
"step": 30379
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 18.0,
|
267 |
+
"learning_rate": 4.4653884628209404e-05,
|
268 |
+
"loss": 3.423,
|
269 |
+
"step": 32166
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 18.0,
|
273 |
+
"eval_accuracy": 0.40659769666211204,
|
274 |
+
"eval_loss": 3.4219350814819336,
|
275 |
+
"eval_runtime": 2.9284,
|
276 |
+
"eval_samples_per_second": 163.569,
|
277 |
+
"eval_steps_per_second": 1.366,
|
278 |
+
"step": 32166
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 19.0,
|
282 |
+
"learning_rate": 4.4355951983994664e-05,
|
283 |
+
"loss": 3.4084,
|
284 |
+
"step": 33953
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 19.0,
|
288 |
+
"eval_accuracy": 0.40138222849083216,
|
289 |
+
"eval_loss": 3.4476888179779053,
|
290 |
+
"eval_runtime": 2.8209,
|
291 |
+
"eval_samples_per_second": 169.805,
|
292 |
+
"eval_steps_per_second": 1.418,
|
293 |
+
"step": 33953
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 20.0,
|
297 |
+
"learning_rate": 4.4058019339779924e-05,
|
298 |
+
"loss": 3.3949,
|
299 |
+
"step": 35740
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 20.0,
|
303 |
+
"eval_accuracy": 0.40872225184519967,
|
304 |
+
"eval_loss": 3.4012813568115234,
|
305 |
+
"eval_runtime": 2.7502,
|
306 |
+
"eval_samples_per_second": 174.171,
|
307 |
+
"eval_steps_per_second": 1.454,
|
308 |
+
"step": 35740
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 21.0,
|
312 |
+
"learning_rate": 4.376008669556519e-05,
|
313 |
+
"loss": 3.3811,
|
314 |
+
"step": 37527
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 21.0,
|
318 |
+
"eval_accuracy": 0.4130312412046158,
|
319 |
+
"eval_loss": 3.3642032146453857,
|
320 |
+
"eval_runtime": 2.8758,
|
321 |
+
"eval_samples_per_second": 166.565,
|
322 |
+
"eval_steps_per_second": 1.391,
|
323 |
+
"step": 37527
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 22.0,
|
327 |
+
"learning_rate": 4.346215405135045e-05,
|
328 |
+
"loss": 3.3688,
|
329 |
+
"step": 39314
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 22.0,
|
333 |
+
"eval_accuracy": 0.403096026951026,
|
334 |
+
"eval_loss": 3.417304277420044,
|
335 |
+
"eval_runtime": 2.7385,
|
336 |
+
"eval_samples_per_second": 174.915,
|
337 |
+
"eval_steps_per_second": 1.461,
|
338 |
+
"step": 39314
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 23.0,
|
342 |
+
"learning_rate": 4.316422140713571e-05,
|
343 |
+
"loss": 3.3598,
|
344 |
+
"step": 41101
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 23.0,
|
348 |
+
"eval_accuracy": 0.4101078204663862,
|
349 |
+
"eval_loss": 3.401829481124878,
|
350 |
+
"eval_runtime": 2.6232,
|
351 |
+
"eval_samples_per_second": 182.602,
|
352 |
+
"eval_steps_per_second": 1.525,
|
353 |
+
"step": 41101
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 24.0,
|
357 |
+
"learning_rate": 4.286628876292098e-05,
|
358 |
+
"loss": 3.3484,
|
359 |
+
"step": 42888
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 24.0,
|
363 |
+
"eval_accuracy": 0.4142865190693482,
|
364 |
+
"eval_loss": 3.3499021530151367,
|
365 |
+
"eval_runtime": 2.8522,
|
366 |
+
"eval_samples_per_second": 167.94,
|
367 |
+
"eval_steps_per_second": 1.402,
|
368 |
+
"step": 42888
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 25.0,
|
372 |
+
"learning_rate": 4.2568356118706237e-05,
|
373 |
+
"loss": 3.3363,
|
374 |
+
"step": 44675
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 25.0,
|
378 |
+
"eval_accuracy": 0.4118788492364802,
|
379 |
+
"eval_loss": 3.3674731254577637,
|
380 |
+
"eval_runtime": 2.7234,
|
381 |
+
"eval_samples_per_second": 175.884,
|
382 |
+
"eval_steps_per_second": 1.469,
|
383 |
+
"step": 44675
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 26.0,
|
387 |
+
"learning_rate": 4.2270423474491496e-05,
|
388 |
+
"loss": 3.3274,
|
389 |
+
"step": 46462
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 26.0,
|
393 |
+
"eval_accuracy": 0.41535205305488926,
|
394 |
+
"eval_loss": 3.356222629547119,
|
395 |
+
"eval_runtime": 2.6874,
|
396 |
+
"eval_samples_per_second": 178.241,
|
397 |
+
"eval_steps_per_second": 1.488,
|
398 |
+
"step": 46462
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 27.0,
|
402 |
+
"learning_rate": 4.197249083027676e-05,
|
403 |
+
"loss": 3.3161,
|
404 |
+
"step": 48249
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 27.0,
|
408 |
+
"eval_accuracy": 0.41593196314670444,
|
409 |
+
"eval_loss": 3.3487422466278076,
|
410 |
+
"eval_runtime": 2.6979,
|
411 |
+
"eval_samples_per_second": 177.545,
|
412 |
+
"eval_steps_per_second": 1.483,
|
413 |
+
"step": 48249
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 28.0,
|
417 |
+
"learning_rate": 4.167455818606202e-05,
|
418 |
+
"loss": 3.3073,
|
419 |
+
"step": 50036
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 28.0,
|
423 |
+
"eval_accuracy": 0.4159050645369928,
|
424 |
+
"eval_loss": 3.329265832901001,
|
425 |
+
"eval_runtime": 2.6601,
|
426 |
+
"eval_samples_per_second": 180.065,
|
427 |
+
"eval_steps_per_second": 1.504,
|
428 |
+
"step": 50036
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 29.0,
|
432 |
+
"learning_rate": 4.137662554184728e-05,
|
433 |
+
"loss": 3.2991,
|
434 |
+
"step": 51823
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 29.0,
|
438 |
+
"eval_accuracy": 0.41597842575425586,
|
439 |
+
"eval_loss": 3.3317177295684814,
|
440 |
+
"eval_runtime": 2.632,
|
441 |
+
"eval_samples_per_second": 181.991,
|
442 |
+
"eval_steps_per_second": 1.52,
|
443 |
+
"step": 51823
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 30.0,
|
447 |
+
"learning_rate": 4.107869289763255e-05,
|
448 |
+
"loss": 3.2899,
|
449 |
+
"step": 53610
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 30.0,
|
453 |
+
"eval_accuracy": 0.4182503044549548,
|
454 |
+
"eval_loss": 3.3058271408081055,
|
455 |
+
"eval_runtime": 2.6549,
|
456 |
+
"eval_samples_per_second": 180.418,
|
457 |
+
"eval_steps_per_second": 1.507,
|
458 |
+
"step": 53610
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 31.0,
|
462 |
+
"learning_rate": 4.078076025341781e-05,
|
463 |
+
"loss": 3.2814,
|
464 |
+
"step": 55397
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 31.0,
|
468 |
+
"eval_accuracy": 0.42345443682608325,
|
469 |
+
"eval_loss": 3.2795028686523438,
|
470 |
+
"eval_runtime": 2.6768,
|
471 |
+
"eval_samples_per_second": 178.945,
|
472 |
+
"eval_steps_per_second": 1.494,
|
473 |
+
"step": 55397
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 32.0,
|
477 |
+
"learning_rate": 4.0482827609203076e-05,
|
478 |
+
"loss": 3.2734,
|
479 |
+
"step": 57184
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 32.0,
|
483 |
+
"eval_accuracy": 0.4142510746169955,
|
484 |
+
"eval_loss": 3.318493366241455,
|
485 |
+
"eval_runtime": 2.6511,
|
486 |
+
"eval_samples_per_second": 180.679,
|
487 |
+
"eval_steps_per_second": 1.509,
|
488 |
+
"step": 57184
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 33.0,
|
492 |
+
"learning_rate": 4.018489496498833e-05,
|
493 |
+
"loss": 3.266,
|
494 |
+
"step": 58971
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 33.0,
|
498 |
+
"eval_accuracy": 0.4268454481298518,
|
499 |
+
"eval_loss": 3.268219232559204,
|
500 |
+
"eval_runtime": 2.8163,
|
501 |
+
"eval_samples_per_second": 170.079,
|
502 |
+
"eval_steps_per_second": 1.42,
|
503 |
+
"step": 58971
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 34.0,
|
507 |
+
"learning_rate": 3.988696232077359e-05,
|
508 |
+
"loss": 3.2578,
|
509 |
+
"step": 60758
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 34.0,
|
513 |
+
"eval_accuracy": 0.4180501984459724,
|
514 |
+
"eval_loss": 3.3144567012786865,
|
515 |
+
"eval_runtime": 2.6553,
|
516 |
+
"eval_samples_per_second": 180.397,
|
517 |
+
"eval_steps_per_second": 1.506,
|
518 |
+
"step": 60758
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 35.0,
|
522 |
+
"learning_rate": 3.9589029676558855e-05,
|
523 |
+
"loss": 3.2506,
|
524 |
+
"step": 62545
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 35.0,
|
528 |
+
"eval_accuracy": 0.423013128672233,
|
529 |
+
"eval_loss": 3.272632598876953,
|
530 |
+
"eval_runtime": 2.6457,
|
531 |
+
"eval_samples_per_second": 181.048,
|
532 |
+
"eval_steps_per_second": 1.512,
|
533 |
+
"step": 62545
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 36.0,
|
537 |
+
"learning_rate": 3.9291097032344115e-05,
|
538 |
+
"loss": 3.2423,
|
539 |
+
"step": 64332
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 36.0,
|
543 |
+
"eval_accuracy": 0.42176717460496294,
|
544 |
+
"eval_loss": 3.273455858230591,
|
545 |
+
"eval_runtime": 2.6217,
|
546 |
+
"eval_samples_per_second": 182.708,
|
547 |
+
"eval_steps_per_second": 1.526,
|
548 |
+
"step": 64332
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 37.0,
|
552 |
+
"learning_rate": 3.8993164388129375e-05,
|
553 |
+
"loss": 3.2359,
|
554 |
+
"step": 66119
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 37.0,
|
558 |
+
"eval_accuracy": 0.417536709284797,
|
559 |
+
"eval_loss": 3.284508228302002,
|
560 |
+
"eval_runtime": 2.7995,
|
561 |
+
"eval_samples_per_second": 171.1,
|
562 |
+
"eval_steps_per_second": 1.429,
|
563 |
+
"step": 66119
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 38.0,
|
567 |
+
"learning_rate": 3.869523174391464e-05,
|
568 |
+
"loss": 3.2293,
|
569 |
+
"step": 67906
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 38.0,
|
573 |
+
"eval_accuracy": 0.4193099997219983,
|
574 |
+
"eval_loss": 3.306713581085205,
|
575 |
+
"eval_runtime": 2.606,
|
576 |
+
"eval_samples_per_second": 183.804,
|
577 |
+
"eval_steps_per_second": 1.535,
|
578 |
+
"step": 67906
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 39.0,
|
582 |
+
"learning_rate": 3.83972990996999e-05,
|
583 |
+
"loss": 3.2207,
|
584 |
+
"step": 69693
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 39.0,
|
588 |
+
"eval_accuracy": 0.42573705623362873,
|
589 |
+
"eval_loss": 3.2586092948913574,
|
590 |
+
"eval_runtime": 2.8824,
|
591 |
+
"eval_samples_per_second": 166.182,
|
592 |
+
"eval_steps_per_second": 1.388,
|
593 |
+
"step": 69693
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 40.0,
|
597 |
+
"learning_rate": 3.809936645548516e-05,
|
598 |
+
"loss": 3.2138,
|
599 |
+
"step": 71480
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 40.0,
|
603 |
+
"eval_accuracy": 0.4249552472588946,
|
604 |
+
"eval_loss": 3.2543227672576904,
|
605 |
+
"eval_runtime": 2.8002,
|
606 |
+
"eval_samples_per_second": 171.061,
|
607 |
+
"eval_steps_per_second": 1.428,
|
608 |
+
"step": 71480
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 41.0,
|
612 |
+
"learning_rate": 3.780143381127043e-05,
|
613 |
+
"loss": 3.2077,
|
614 |
+
"step": 73267
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 41.0,
|
618 |
+
"eval_accuracy": 0.42259816679868734,
|
619 |
+
"eval_loss": 3.239523410797119,
|
620 |
+
"eval_runtime": 2.8097,
|
621 |
+
"eval_samples_per_second": 170.48,
|
622 |
+
"eval_steps_per_second": 1.424,
|
623 |
+
"step": 73267
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 42.0,
|
627 |
+
"learning_rate": 3.750350116705569e-05,
|
628 |
+
"loss": 3.202,
|
629 |
+
"step": 75054
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 42.0,
|
633 |
+
"eval_accuracy": 0.4270068903702874,
|
634 |
+
"eval_loss": 3.222418785095215,
|
635 |
+
"eval_runtime": 2.7468,
|
636 |
+
"eval_samples_per_second": 174.388,
|
637 |
+
"eval_steps_per_second": 1.456,
|
638 |
+
"step": 75054
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 43.0,
|
642 |
+
"learning_rate": 3.720556852284095e-05,
|
643 |
+
"loss": 3.1964,
|
644 |
+
"step": 76841
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 43.0,
|
648 |
+
"eval_accuracy": 0.4233582733012544,
|
649 |
+
"eval_loss": 3.256213426589966,
|
650 |
+
"eval_runtime": 2.7921,
|
651 |
+
"eval_samples_per_second": 171.556,
|
652 |
+
"eval_steps_per_second": 1.433,
|
653 |
+
"step": 76841
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 44.0,
|
657 |
+
"learning_rate": 3.6907635878626214e-05,
|
658 |
+
"loss": 3.1925,
|
659 |
+
"step": 78628
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 44.0,
|
663 |
+
"eval_accuracy": 0.4251221214235869,
|
664 |
+
"eval_loss": 3.2543723583221436,
|
665 |
+
"eval_runtime": 2.7441,
|
666 |
+
"eval_samples_per_second": 174.557,
|
667 |
+
"eval_steps_per_second": 1.458,
|
668 |
+
"step": 78628
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 45.0,
|
672 |
+
"learning_rate": 3.660970323441147e-05,
|
673 |
+
"loss": 3.1865,
|
674 |
+
"step": 80415
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 45.0,
|
678 |
+
"eval_accuracy": 0.43534812547819435,
|
679 |
+
"eval_loss": 3.204282522201538,
|
680 |
+
"eval_runtime": 2.8747,
|
681 |
+
"eval_samples_per_second": 166.629,
|
682 |
+
"eval_steps_per_second": 1.391,
|
683 |
+
"step": 80415
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 46.0,
|
687 |
+
"learning_rate": 3.631177059019673e-05,
|
688 |
+
"loss": 3.1812,
|
689 |
+
"step": 82202
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 46.0,
|
693 |
+
"eval_accuracy": 0.42858336117607754,
|
694 |
+
"eval_loss": 3.2280006408691406,
|
695 |
+
"eval_runtime": 2.742,
|
696 |
+
"eval_samples_per_second": 174.687,
|
697 |
+
"eval_steps_per_second": 1.459,
|
698 |
+
"step": 82202
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 47.0,
|
702 |
+
"learning_rate": 3.601383794598199e-05,
|
703 |
+
"loss": 3.1744,
|
704 |
+
"step": 83989
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 47.0,
|
708 |
+
"eval_accuracy": 0.427574415606256,
|
709 |
+
"eval_loss": 3.21743106842041,
|
710 |
+
"eval_runtime": 2.7215,
|
711 |
+
"eval_samples_per_second": 176.009,
|
712 |
+
"eval_steps_per_second": 1.47,
|
713 |
+
"step": 83989
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 48.0,
|
717 |
+
"learning_rate": 3.571590530176725e-05,
|
718 |
+
"loss": 3.1699,
|
719 |
+
"step": 85776
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 48.0,
|
723 |
+
"eval_accuracy": 0.4317117218804673,
|
724 |
+
"eval_loss": 3.1971514225006104,
|
725 |
+
"eval_runtime": 2.9267,
|
726 |
+
"eval_samples_per_second": 163.664,
|
727 |
+
"eval_steps_per_second": 1.367,
|
728 |
+
"step": 85776
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 49.0,
|
732 |
+
"learning_rate": 3.541797265755252e-05,
|
733 |
+
"loss": 3.1652,
|
734 |
+
"step": 87563
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 49.0,
|
738 |
+
"eval_accuracy": 0.43024569330697543,
|
739 |
+
"eval_loss": 3.201594114303589,
|
740 |
+
"eval_runtime": 2.8821,
|
741 |
+
"eval_samples_per_second": 166.197,
|
742 |
+
"eval_steps_per_second": 1.388,
|
743 |
+
"step": 87563
|
744 |
+
},
|
745 |
+
{
|
746 |
+
"epoch": 50.0,
|
747 |
+
"learning_rate": 3.512004001333778e-05,
|
748 |
+
"loss": 3.1609,
|
749 |
+
"step": 89350
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 50.0,
|
753 |
+
"eval_accuracy": 0.4337845459672871,
|
754 |
+
"eval_loss": 3.201768636703491,
|
755 |
+
"eval_runtime": 2.7518,
|
756 |
+
"eval_samples_per_second": 174.066,
|
757 |
+
"eval_steps_per_second": 1.454,
|
758 |
+
"step": 89350
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 51.0,
|
762 |
+
"learning_rate": 3.482210736912304e-05,
|
763 |
+
"loss": 3.1548,
|
764 |
+
"step": 91137
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 51.0,
|
768 |
+
"eval_accuracy": 0.43272542487719495,
|
769 |
+
"eval_loss": 3.1950149536132812,
|
770 |
+
"eval_runtime": 2.6564,
|
771 |
+
"eval_samples_per_second": 180.321,
|
772 |
+
"eval_steps_per_second": 1.506,
|
773 |
+
"step": 91137
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 52.0,
|
777 |
+
"learning_rate": 3.4524174724908306e-05,
|
778 |
+
"loss": 3.1508,
|
779 |
+
"step": 92924
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 52.0,
|
783 |
+
"eval_accuracy": 0.42785334961519894,
|
784 |
+
"eval_loss": 3.212803363800049,
|
785 |
+
"eval_runtime": 2.8096,
|
786 |
+
"eval_samples_per_second": 170.486,
|
787 |
+
"eval_steps_per_second": 1.424,
|
788 |
+
"step": 92924
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 53.0,
|
792 |
+
"learning_rate": 3.4226242080693565e-05,
|
793 |
+
"loss": 3.1478,
|
794 |
+
"step": 94711
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 53.0,
|
798 |
+
"eval_accuracy": 0.4303176427710337,
|
799 |
+
"eval_loss": 3.202669382095337,
|
800 |
+
"eval_runtime": 2.727,
|
801 |
+
"eval_samples_per_second": 175.65,
|
802 |
+
"eval_steps_per_second": 1.467,
|
803 |
+
"step": 94711
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 54.0,
|
807 |
+
"learning_rate": 3.3928309436478825e-05,
|
808 |
+
"loss": 3.1423,
|
809 |
+
"step": 96498
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 54.0,
|
813 |
+
"eval_accuracy": 0.43117465657415194,
|
814 |
+
"eval_loss": 3.195861577987671,
|
815 |
+
"eval_runtime": 2.7557,
|
816 |
+
"eval_samples_per_second": 173.823,
|
817 |
+
"eval_steps_per_second": 1.452,
|
818 |
+
"step": 96498
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 55.0,
|
822 |
+
"learning_rate": 3.363037679226409e-05,
|
823 |
+
"loss": 3.1383,
|
824 |
+
"step": 98285
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 55.0,
|
828 |
+
"eval_accuracy": 0.4340061025109873,
|
829 |
+
"eval_loss": 3.1911423206329346,
|
830 |
+
"eval_runtime": 2.754,
|
831 |
+
"eval_samples_per_second": 173.929,
|
832 |
+
"eval_steps_per_second": 1.452,
|
833 |
+
"step": 98285
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 56.0,
|
837 |
+
"learning_rate": 3.333244414804935e-05,
|
838 |
+
"loss": 3.1336,
|
839 |
+
"step": 100072
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 56.0,
|
843 |
+
"eval_accuracy": 0.43203597021216805,
|
844 |
+
"eval_loss": 3.1913583278656006,
|
845 |
+
"eval_runtime": 2.6171,
|
846 |
+
"eval_samples_per_second": 183.026,
|
847 |
+
"eval_steps_per_second": 1.528,
|
848 |
+
"step": 100072
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 57.0,
|
852 |
+
"learning_rate": 3.303451150383461e-05,
|
853 |
+
"loss": 3.129,
|
854 |
+
"step": 101859
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 57.0,
|
858 |
+
"eval_accuracy": 0.43117934403349617,
|
859 |
+
"eval_loss": 3.185468912124634,
|
860 |
+
"eval_runtime": 2.8993,
|
861 |
+
"eval_samples_per_second": 165.21,
|
862 |
+
"eval_steps_per_second": 1.38,
|
863 |
+
"step": 101859
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 58.0,
|
867 |
+
"learning_rate": 3.273657885961988e-05,
|
868 |
+
"loss": 3.1233,
|
869 |
+
"step": 103646
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 58.0,
|
873 |
+
"eval_accuracy": 0.4336863338376198,
|
874 |
+
"eval_loss": 3.1569876670837402,
|
875 |
+
"eval_runtime": 2.7946,
|
876 |
+
"eval_samples_per_second": 171.403,
|
877 |
+
"eval_steps_per_second": 1.431,
|
878 |
+
"step": 103646
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 59.0,
|
882 |
+
"learning_rate": 3.243864621540514e-05,
|
883 |
+
"loss": 3.1198,
|
884 |
+
"step": 105433
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 59.0,
|
888 |
+
"eval_accuracy": 0.4307147438578878,
|
889 |
+
"eval_loss": 3.2042369842529297,
|
890 |
+
"eval_runtime": 2.9712,
|
891 |
+
"eval_samples_per_second": 161.213,
|
892 |
+
"eval_steps_per_second": 1.346,
|
893 |
+
"step": 105433
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 60.0,
|
897 |
+
"learning_rate": 3.21407135711904e-05,
|
898 |
+
"loss": 3.1153,
|
899 |
+
"step": 107220
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 60.0,
|
903 |
+
"eval_accuracy": 0.4389830036996857,
|
904 |
+
"eval_loss": 3.1370162963867188,
|
905 |
+
"eval_runtime": 2.6328,
|
906 |
+
"eval_samples_per_second": 181.937,
|
907 |
+
"eval_steps_per_second": 1.519,
|
908 |
+
"step": 107220
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 61.0,
|
912 |
+
"learning_rate": 3.1842780926975664e-05,
|
913 |
+
"loss": 3.1122,
|
914 |
+
"step": 109007
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 61.0,
|
918 |
+
"eval_accuracy": 0.4412110687878109,
|
919 |
+
"eval_loss": 3.1612205505371094,
|
920 |
+
"eval_runtime": 2.8382,
|
921 |
+
"eval_samples_per_second": 168.771,
|
922 |
+
"eval_steps_per_second": 1.409,
|
923 |
+
"step": 109007
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 62.0,
|
927 |
+
"learning_rate": 3.154484828276092e-05,
|
928 |
+
"loss": 3.1093,
|
929 |
+
"step": 110794
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 62.0,
|
933 |
+
"eval_accuracy": 0.4347947500698129,
|
934 |
+
"eval_loss": 3.164184331893921,
|
935 |
+
"eval_runtime": 2.8661,
|
936 |
+
"eval_samples_per_second": 167.128,
|
937 |
+
"eval_steps_per_second": 1.396,
|
938 |
+
"step": 110794
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 63.0,
|
942 |
+
"learning_rate": 3.1246915638546184e-05,
|
943 |
+
"loss": 3.1048,
|
944 |
+
"step": 112581
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 63.0,
|
948 |
+
"eval_accuracy": 0.4325971058644326,
|
949 |
+
"eval_loss": 3.1806650161743164,
|
950 |
+
"eval_runtime": 2.8065,
|
951 |
+
"eval_samples_per_second": 170.676,
|
952 |
+
"eval_steps_per_second": 1.425,
|
953 |
+
"step": 112581
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 64.0,
|
957 |
+
"learning_rate": 3.0948982994331444e-05,
|
958 |
+
"loss": 3.1013,
|
959 |
+
"step": 114368
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 64.0,
|
963 |
+
"eval_accuracy": 0.4358816103240286,
|
964 |
+
"eval_loss": 3.144869089126587,
|
965 |
+
"eval_runtime": 2.6572,
|
966 |
+
"eval_samples_per_second": 180.267,
|
967 |
+
"eval_steps_per_second": 1.505,
|
968 |
+
"step": 114368
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 65.0,
|
972 |
+
"learning_rate": 3.06510503501167e-05,
|
973 |
+
"loss": 3.0977,
|
974 |
+
"step": 116155
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 65.0,
|
978 |
+
"eval_accuracy": 0.43795185846729606,
|
979 |
+
"eval_loss": 3.1408073902130127,
|
980 |
+
"eval_runtime": 2.8767,
|
981 |
+
"eval_samples_per_second": 166.508,
|
982 |
+
"eval_steps_per_second": 1.39,
|
983 |
+
"step": 116155
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 66.0,
|
987 |
+
"learning_rate": 3.0353117705901967e-05,
|
988 |
+
"loss": 3.0926,
|
989 |
+
"step": 117942
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 66.0,
|
993 |
+
"eval_accuracy": 0.43651985318651987,
|
994 |
+
"eval_loss": 3.172287702560425,
|
995 |
+
"eval_runtime": 2.6811,
|
996 |
+
"eval_samples_per_second": 178.655,
|
997 |
+
"eval_steps_per_second": 1.492,
|
998 |
+
"step": 117942
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 67.0,
|
1002 |
+
"learning_rate": 3.005518506168723e-05,
|
1003 |
+
"loss": 3.0901,
|
1004 |
+
"step": 119729
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 67.0,
|
1008 |
+
"eval_accuracy": 0.43799220873854433,
|
1009 |
+
"eval_loss": 3.1473004817962646,
|
1010 |
+
"eval_runtime": 2.8395,
|
1011 |
+
"eval_samples_per_second": 168.689,
|
1012 |
+
"eval_steps_per_second": 1.409,
|
1013 |
+
"step": 119729
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 68.0,
|
1017 |
+
"learning_rate": 2.9757252417472493e-05,
|
1018 |
+
"loss": 3.0882,
|
1019 |
+
"step": 121516
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 68.0,
|
1023 |
+
"eval_accuracy": 0.43776389123458875,
|
1024 |
+
"eval_loss": 3.140061378479004,
|
1025 |
+
"eval_runtime": 2.8036,
|
1026 |
+
"eval_samples_per_second": 170.853,
|
1027 |
+
"eval_steps_per_second": 1.427,
|
1028 |
+
"step": 121516
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 69.0,
|
1032 |
+
"learning_rate": 2.9459319773257753e-05,
|
1033 |
+
"loss": 3.0839,
|
1034 |
+
"step": 123303
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 69.0,
|
1038 |
+
"eval_accuracy": 0.43737838066499335,
|
1039 |
+
"eval_loss": 3.12809157371521,
|
1040 |
+
"eval_runtime": 2.7834,
|
1041 |
+
"eval_samples_per_second": 172.089,
|
1042 |
+
"eval_steps_per_second": 1.437,
|
1043 |
+
"step": 123303
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 70.0,
|
1047 |
+
"learning_rate": 2.9161387129043016e-05,
|
1048 |
+
"loss": 3.0794,
|
1049 |
+
"step": 125090
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 70.0,
|
1053 |
+
"eval_accuracy": 0.436696631101403,
|
1054 |
+
"eval_loss": 3.1356077194213867,
|
1055 |
+
"eval_runtime": 2.7463,
|
1056 |
+
"eval_samples_per_second": 174.417,
|
1057 |
+
"eval_steps_per_second": 1.457,
|
1058 |
+
"step": 125090
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 71.0,
|
1062 |
+
"learning_rate": 2.886345448482828e-05,
|
1063 |
+
"loss": 3.0766,
|
1064 |
+
"step": 126877
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 71.0,
|
1068 |
+
"eval_accuracy": 0.4397433382835047,
|
1069 |
+
"eval_loss": 3.1019225120544434,
|
1070 |
+
"eval_runtime": 2.7446,
|
1071 |
+
"eval_samples_per_second": 174.525,
|
1072 |
+
"eval_steps_per_second": 1.457,
|
1073 |
+
"step": 126877
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 72.0,
|
1077 |
+
"learning_rate": 2.856552184061354e-05,
|
1078 |
+
"loss": 3.074,
|
1079 |
+
"step": 128664
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 72.0,
|
1083 |
+
"eval_accuracy": 0.43546703529212627,
|
1084 |
+
"eval_loss": 3.162637948989868,
|
1085 |
+
"eval_runtime": 2.7735,
|
1086 |
+
"eval_samples_per_second": 172.704,
|
1087 |
+
"eval_steps_per_second": 1.442,
|
1088 |
+
"step": 128664
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 73.0,
|
1092 |
+
"learning_rate": 2.8267589196398802e-05,
|
1093 |
+
"loss": 3.0702,
|
1094 |
+
"step": 130451
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 73.0,
|
1098 |
+
"eval_accuracy": 0.4387263241465351,
|
1099 |
+
"eval_loss": 3.1287012100219727,
|
1100 |
+
"eval_runtime": 2.7531,
|
1101 |
+
"eval_samples_per_second": 173.986,
|
1102 |
+
"eval_steps_per_second": 1.453,
|
1103 |
+
"step": 130451
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 74.0,
|
1107 |
+
"learning_rate": 2.7969656552184065e-05,
|
1108 |
+
"loss": 3.0676,
|
1109 |
+
"step": 132238
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 74.0,
|
1113 |
+
"eval_accuracy": 0.43789707310306564,
|
1114 |
+
"eval_loss": 3.136582136154175,
|
1115 |
+
"eval_runtime": 2.7895,
|
1116 |
+
"eval_samples_per_second": 171.718,
|
1117 |
+
"eval_steps_per_second": 1.434,
|
1118 |
+
"step": 132238
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 75.0,
|
1122 |
+
"learning_rate": 2.767172390796933e-05,
|
1123 |
+
"loss": 3.0648,
|
1124 |
+
"step": 134025
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 75.0,
|
1128 |
+
"eval_accuracy": 0.43458386755114553,
|
1129 |
+
"eval_loss": 3.178231716156006,
|
1130 |
+
"eval_runtime": 2.7934,
|
1131 |
+
"eval_samples_per_second": 171.478,
|
1132 |
+
"eval_steps_per_second": 1.432,
|
1133 |
+
"step": 134025
|
1134 |
+
},
|
1135 |
+
{
|
1136 |
+
"epoch": 76.0,
|
1137 |
+
"learning_rate": 2.7373791263754585e-05,
|
1138 |
+
"loss": 3.0624,
|
1139 |
+
"step": 135812
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 76.0,
|
1143 |
+
"eval_accuracy": 0.44267685994547346,
|
1144 |
+
"eval_loss": 3.1229066848754883,
|
1145 |
+
"eval_runtime": 2.7648,
|
1146 |
+
"eval_samples_per_second": 173.25,
|
1147 |
+
"eval_steps_per_second": 1.447,
|
1148 |
+
"step": 135812
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 77.0,
|
1152 |
+
"learning_rate": 2.7075858619539845e-05,
|
1153 |
+
"loss": 3.0575,
|
1154 |
+
"step": 137599
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 77.0,
|
1158 |
+
"eval_accuracy": 0.44304284497210544,
|
1159 |
+
"eval_loss": 3.11391282081604,
|
1160 |
+
"eval_runtime": 2.8016,
|
1161 |
+
"eval_samples_per_second": 170.976,
|
1162 |
+
"eval_steps_per_second": 1.428,
|
1163 |
+
"step": 137599
|
1164 |
+
},
|
1165 |
+
{
|
1166 |
+
"epoch": 78.0,
|
1167 |
+
"learning_rate": 2.6777925975325108e-05,
|
1168 |
+
"loss": 3.0549,
|
1169 |
+
"step": 139386
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 78.0,
|
1173 |
+
"eval_accuracy": 0.4431432021793468,
|
1174 |
+
"eval_loss": 3.0947771072387695,
|
1175 |
+
"eval_runtime": 2.8076,
|
1176 |
+
"eval_samples_per_second": 170.609,
|
1177 |
+
"eval_steps_per_second": 1.425,
|
1178 |
+
"step": 139386
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 79.0,
|
1182 |
+
"learning_rate": 2.647999333111037e-05,
|
1183 |
+
"loss": 3.052,
|
1184 |
+
"step": 141173
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 79.0,
|
1188 |
+
"eval_accuracy": 0.4451516602360098,
|
1189 |
+
"eval_loss": 3.1029744148254395,
|
1190 |
+
"eval_runtime": 2.6939,
|
1191 |
+
"eval_samples_per_second": 177.81,
|
1192 |
+
"eval_steps_per_second": 1.485,
|
1193 |
+
"step": 141173
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 80.0,
|
1197 |
+
"learning_rate": 2.618206068689563e-05,
|
1198 |
+
"loss": 3.0527,
|
1199 |
+
"step": 142960
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 80.0,
|
1203 |
+
"eval_accuracy": 0.4447775411856999,
|
1204 |
+
"eval_loss": 3.0929136276245117,
|
1205 |
+
"eval_runtime": 2.7294,
|
1206 |
+
"eval_samples_per_second": 175.498,
|
1207 |
+
"eval_steps_per_second": 1.466,
|
1208 |
+
"step": 142960
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 81.0,
|
1212 |
+
"learning_rate": 2.5884128042680894e-05,
|
1213 |
+
"loss": 3.0466,
|
1214 |
+
"step": 144747
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 81.0,
|
1218 |
+
"eval_accuracy": 0.44276007765216396,
|
1219 |
+
"eval_loss": 3.08884334564209,
|
1220 |
+
"eval_runtime": 2.9041,
|
1221 |
+
"eval_samples_per_second": 164.938,
|
1222 |
+
"eval_steps_per_second": 1.377,
|
1223 |
+
"step": 144747
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 82.0,
|
1227 |
+
"learning_rate": 2.5586195398466157e-05,
|
1228 |
+
"loss": 3.0439,
|
1229 |
+
"step": 146534
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 82.0,
|
1233 |
+
"eval_accuracy": 0.44138026993182133,
|
1234 |
+
"eval_loss": 3.103548765182495,
|
1235 |
+
"eval_runtime": 2.7889,
|
1236 |
+
"eval_samples_per_second": 171.752,
|
1237 |
+
"eval_steps_per_second": 1.434,
|
1238 |
+
"step": 146534
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 83.0,
|
1242 |
+
"learning_rate": 2.5288262754251417e-05,
|
1243 |
+
"loss": 3.0409,
|
1244 |
+
"step": 148321
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 83.0,
|
1248 |
+
"eval_accuracy": 0.4411418975650714,
|
1249 |
+
"eval_loss": 3.1111607551574707,
|
1250 |
+
"eval_runtime": 2.7759,
|
1251 |
+
"eval_samples_per_second": 172.554,
|
1252 |
+
"eval_steps_per_second": 1.441,
|
1253 |
+
"step": 148321
|
1254 |
+
},
|
1255 |
+
{
|
1256 |
+
"epoch": 84.0,
|
1257 |
+
"learning_rate": 2.499033011003668e-05,
|
1258 |
+
"loss": 3.041,
|
1259 |
+
"step": 150108
|
1260 |
+
},
|
1261 |
+
{
|
1262 |
+
"epoch": 84.0,
|
1263 |
+
"eval_accuracy": 0.43987652588747017,
|
1264 |
+
"eval_loss": 3.129586696624756,
|
1265 |
+
"eval_runtime": 2.8127,
|
1266 |
+
"eval_samples_per_second": 170.3,
|
1267 |
+
"eval_steps_per_second": 1.422,
|
1268 |
+
"step": 150108
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 85.0,
|
1272 |
+
"learning_rate": 2.4692397465821944e-05,
|
1273 |
+
"loss": 3.0379,
|
1274 |
+
"step": 151895
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 85.0,
|
1278 |
+
"eval_accuracy": 0.44284645440186216,
|
1279 |
+
"eval_loss": 3.122410297393799,
|
1280 |
+
"eval_runtime": 2.8313,
|
1281 |
+
"eval_samples_per_second": 169.183,
|
1282 |
+
"eval_steps_per_second": 1.413,
|
1283 |
+
"step": 151895
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 86.0,
|
1287 |
+
"learning_rate": 2.4394464821607203e-05,
|
1288 |
+
"loss": 3.0332,
|
1289 |
+
"step": 153682
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 86.0,
|
1293 |
+
"eval_accuracy": 0.43977575332866153,
|
1294 |
+
"eval_loss": 3.1100828647613525,
|
1295 |
+
"eval_runtime": 2.7772,
|
1296 |
+
"eval_samples_per_second": 172.476,
|
1297 |
+
"eval_steps_per_second": 1.44,
|
1298 |
+
"step": 153682
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 87.0,
|
1302 |
+
"learning_rate": 2.4096532177392463e-05,
|
1303 |
+
"loss": 3.0315,
|
1304 |
+
"step": 155469
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 87.0,
|
1308 |
+
"eval_accuracy": 0.4423447851731753,
|
1309 |
+
"eval_loss": 3.1045453548431396,
|
1310 |
+
"eval_runtime": 2.6529,
|
1311 |
+
"eval_samples_per_second": 180.554,
|
1312 |
+
"eval_steps_per_second": 1.508,
|
1313 |
+
"step": 155469
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 88.0,
|
1317 |
+
"learning_rate": 2.3798599533177726e-05,
|
1318 |
+
"loss": 3.0302,
|
1319 |
+
"step": 157256
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 88.0,
|
1323 |
+
"eval_accuracy": 0.4446285634040163,
|
1324 |
+
"eval_loss": 3.0912954807281494,
|
1325 |
+
"eval_runtime": 2.7592,
|
1326 |
+
"eval_samples_per_second": 173.599,
|
1327 |
+
"eval_steps_per_second": 1.45,
|
1328 |
+
"step": 157256
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 89.0,
|
1332 |
+
"learning_rate": 2.350066688896299e-05,
|
1333 |
+
"loss": 3.0265,
|
1334 |
+
"step": 159043
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 89.0,
|
1338 |
+
"eval_accuracy": 0.44469112504856523,
|
1339 |
+
"eval_loss": 3.074544906616211,
|
1340 |
+
"eval_runtime": 2.879,
|
1341 |
+
"eval_samples_per_second": 166.379,
|
1342 |
+
"eval_steps_per_second": 1.389,
|
1343 |
+
"step": 159043
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 90.0,
|
1347 |
+
"learning_rate": 2.320273424474825e-05,
|
1348 |
+
"loss": 3.0243,
|
1349 |
+
"step": 160830
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 90.0,
|
1353 |
+
"eval_accuracy": 0.4443120079831467,
|
1354 |
+
"eval_loss": 3.0942282676696777,
|
1355 |
+
"eval_runtime": 2.7539,
|
1356 |
+
"eval_samples_per_second": 173.935,
|
1357 |
+
"eval_steps_per_second": 1.452,
|
1358 |
+
"step": 160830
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 91.0,
|
1362 |
+
"learning_rate": 2.2904801600533513e-05,
|
1363 |
+
"loss": 3.0222,
|
1364 |
+
"step": 162617
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 91.0,
|
1368 |
+
"eval_accuracy": 0.4432415712960099,
|
1369 |
+
"eval_loss": 3.0821101665496826,
|
1370 |
+
"eval_runtime": 2.9525,
|
1371 |
+
"eval_samples_per_second": 162.233,
|
1372 |
+
"eval_steps_per_second": 1.355,
|
1373 |
+
"step": 162617
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 92.0,
|
1377 |
+
"learning_rate": 2.2606868956318776e-05,
|
1378 |
+
"loss": 3.021,
|
1379 |
+
"step": 164404
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 92.0,
|
1383 |
+
"eval_accuracy": 0.4472506123828026,
|
1384 |
+
"eval_loss": 3.0616304874420166,
|
1385 |
+
"eval_runtime": 2.7386,
|
1386 |
+
"eval_samples_per_second": 174.904,
|
1387 |
+
"eval_steps_per_second": 1.461,
|
1388 |
+
"step": 164404
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 93.0,
|
1392 |
+
"learning_rate": 2.2308936312104036e-05,
|
1393 |
+
"loss": 3.0183,
|
1394 |
+
"step": 166191
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 93.0,
|
1398 |
+
"eval_accuracy": 0.4450408140445041,
|
1399 |
+
"eval_loss": 3.102149724960327,
|
1400 |
+
"eval_runtime": 2.7638,
|
1401 |
+
"eval_samples_per_second": 173.309,
|
1402 |
+
"eval_steps_per_second": 1.447,
|
1403 |
+
"step": 166191
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 94.0,
|
1407 |
+
"learning_rate": 2.2011003667889295e-05,
|
1408 |
+
"loss": 3.0155,
|
1409 |
+
"step": 167978
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 94.0,
|
1413 |
+
"eval_accuracy": 0.44215097425485717,
|
1414 |
+
"eval_loss": 3.116255283355713,
|
1415 |
+
"eval_runtime": 2.7634,
|
1416 |
+
"eval_samples_per_second": 173.336,
|
1417 |
+
"eval_steps_per_second": 1.447,
|
1418 |
+
"step": 167978
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 95.0,
|
1422 |
+
"learning_rate": 2.171307102367456e-05,
|
1423 |
+
"loss": 3.0132,
|
1424 |
+
"step": 169765
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 95.0,
|
1428 |
+
"eval_accuracy": 0.4493496925561758,
|
1429 |
+
"eval_loss": 3.0645430088043213,
|
1430 |
+
"eval_runtime": 2.9971,
|
1431 |
+
"eval_samples_per_second": 159.823,
|
1432 |
+
"eval_steps_per_second": 1.335,
|
1433 |
+
"step": 169765
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 96.0,
|
1437 |
+
"learning_rate": 2.1415138379459822e-05,
|
1438 |
+
"loss": 3.0118,
|
1439 |
+
"step": 171552
|
1440 |
+
},
|
1441 |
+
{
|
1442 |
+
"epoch": 96.0,
|
1443 |
+
"eval_accuracy": 0.44202857943401513,
|
1444 |
+
"eval_loss": 3.092226982116699,
|
1445 |
+
"eval_runtime": 2.7558,
|
1446 |
+
"eval_samples_per_second": 173.815,
|
1447 |
+
"eval_steps_per_second": 1.451,
|
1448 |
+
"step": 171552
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 97.0,
|
1452 |
+
"learning_rate": 2.111720573524508e-05,
|
1453 |
+
"loss": 3.0105,
|
1454 |
+
"step": 173339
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 97.0,
|
1458 |
+
"eval_accuracy": 0.4422707131510743,
|
1459 |
+
"eval_loss": 3.118738889694214,
|
1460 |
+
"eval_runtime": 2.8983,
|
1461 |
+
"eval_samples_per_second": 165.268,
|
1462 |
+
"eval_steps_per_second": 1.38,
|
1463 |
+
"step": 173339
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 98.0,
|
1467 |
+
"learning_rate": 2.0819273091030345e-05,
|
1468 |
+
"loss": 3.0063,
|
1469 |
+
"step": 175126
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 98.0,
|
1473 |
+
"eval_accuracy": 0.44618075143856484,
|
1474 |
+
"eval_loss": 3.1060523986816406,
|
1475 |
+
"eval_runtime": 2.8061,
|
1476 |
+
"eval_samples_per_second": 170.697,
|
1477 |
+
"eval_steps_per_second": 1.425,
|
1478 |
+
"step": 175126
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 99.0,
|
1482 |
+
"learning_rate": 2.0521340446815608e-05,
|
1483 |
+
"loss": 3.0035,
|
1484 |
+
"step": 176913
|
1485 |
+
},
|
1486 |
+
{
|
1487 |
+
"epoch": 99.0,
|
1488 |
+
"eval_accuracy": 0.4423852183650616,
|
1489 |
+
"eval_loss": 3.1097826957702637,
|
1490 |
+
"eval_runtime": 3.1412,
|
1491 |
+
"eval_samples_per_second": 152.491,
|
1492 |
+
"eval_steps_per_second": 1.273,
|
1493 |
+
"step": 176913
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 100.0,
|
1497 |
+
"learning_rate": 2.0223407802600868e-05,
|
1498 |
+
"loss": 3.0025,
|
1499 |
+
"step": 178700
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 100.0,
|
1503 |
+
"eval_accuracy": 0.4454137587238285,
|
1504 |
+
"eval_loss": 3.0856051445007324,
|
1505 |
+
"eval_runtime": 2.8018,
|
1506 |
+
"eval_samples_per_second": 170.961,
|
1507 |
+
"eval_steps_per_second": 1.428,
|
1508 |
+
"step": 178700
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 101.0,
|
1512 |
+
"learning_rate": 1.992547515838613e-05,
|
1513 |
+
"loss": 3.0001,
|
1514 |
+
"step": 180487
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 101.0,
|
1518 |
+
"eval_accuracy": 0.4503842392063714,
|
1519 |
+
"eval_loss": 3.0584394931793213,
|
1520 |
+
"eval_runtime": 2.8738,
|
1521 |
+
"eval_samples_per_second": 166.676,
|
1522 |
+
"eval_steps_per_second": 1.392,
|
1523 |
+
"step": 180487
|
1524 |
+
},
|
1525 |
+
{
|
1526 |
+
"epoch": 102.0,
|
1527 |
+
"learning_rate": 1.962754251417139e-05,
|
1528 |
+
"loss": 2.9979,
|
1529 |
+
"step": 182274
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 102.0,
|
1533 |
+
"eval_accuracy": 0.4435431811165062,
|
1534 |
+
"eval_loss": 3.089740753173828,
|
1535 |
+
"eval_runtime": 3.0249,
|
1536 |
+
"eval_samples_per_second": 158.354,
|
1537 |
+
"eval_steps_per_second": 1.322,
|
1538 |
+
"step": 182274
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 103.0,
|
1542 |
+
"learning_rate": 1.9329609869956654e-05,
|
1543 |
+
"loss": 2.9963,
|
1544 |
+
"step": 184061
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 103.0,
|
1548 |
+
"eval_accuracy": 0.4437190013170062,
|
1549 |
+
"eval_loss": 3.0712223052978516,
|
1550 |
+
"eval_runtime": 2.8272,
|
1551 |
+
"eval_samples_per_second": 169.429,
|
1552 |
+
"eval_steps_per_second": 1.415,
|
1553 |
+
"step": 184061
|
1554 |
+
},
|
1555 |
+
{
|
1556 |
+
"epoch": 104.0,
|
1557 |
+
"learning_rate": 1.9031677225741914e-05,
|
1558 |
+
"loss": 2.9944,
|
1559 |
+
"step": 185848
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 104.0,
|
1563 |
+
"eval_accuracy": 0.445785838620922,
|
1564 |
+
"eval_loss": 3.085341453552246,
|
1565 |
+
"eval_runtime": 2.8252,
|
1566 |
+
"eval_samples_per_second": 169.547,
|
1567 |
+
"eval_steps_per_second": 1.416,
|
1568 |
+
"step": 185848
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 105.0,
|
1572 |
+
"learning_rate": 1.8733744581527177e-05,
|
1573 |
+
"loss": 2.9931,
|
1574 |
+
"step": 187635
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 105.0,
|
1578 |
+
"eval_accuracy": 0.44751641410643545,
|
1579 |
+
"eval_loss": 3.0809359550476074,
|
1580 |
+
"eval_runtime": 2.5855,
|
1581 |
+
"eval_samples_per_second": 185.266,
|
1582 |
+
"eval_steps_per_second": 1.547,
|
1583 |
+
"step": 187635
|
1584 |
+
},
|
1585 |
+
{
|
1586 |
+
"epoch": 106.0,
|
1587 |
+
"learning_rate": 1.843581193731244e-05,
|
1588 |
+
"loss": 2.992,
|
1589 |
+
"step": 189422
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 106.0,
|
1593 |
+
"eval_accuracy": 0.44257175014068656,
|
1594 |
+
"eval_loss": 3.0910277366638184,
|
1595 |
+
"eval_runtime": 2.6969,
|
1596 |
+
"eval_samples_per_second": 177.61,
|
1597 |
+
"eval_steps_per_second": 1.483,
|
1598 |
+
"step": 189422
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 107.0,
|
1602 |
+
"learning_rate": 1.81378792930977e-05,
|
1603 |
+
"loss": 2.9886,
|
1604 |
+
"step": 191209
|
1605 |
+
},
|
1606 |
+
{
|
1607 |
+
"epoch": 107.0,
|
1608 |
+
"eval_accuracy": 0.44900508104118764,
|
1609 |
+
"eval_loss": 3.0693163871765137,
|
1610 |
+
"eval_runtime": 2.7589,
|
1611 |
+
"eval_samples_per_second": 173.62,
|
1612 |
+
"eval_steps_per_second": 1.45,
|
1613 |
+
"step": 191209
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 108.0,
|
1617 |
+
"learning_rate": 1.7839946648882963e-05,
|
1618 |
+
"loss": 2.986,
|
1619 |
+
"step": 192996
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 108.0,
|
1623 |
+
"eval_accuracy": 0.4444630171333055,
|
1624 |
+
"eval_loss": 3.0906262397766113,
|
1625 |
+
"eval_runtime": 3.2348,
|
1626 |
+
"eval_samples_per_second": 148.078,
|
1627 |
+
"eval_steps_per_second": 1.237,
|
1628 |
+
"step": 192996
|
1629 |
+
},
|
1630 |
+
{
|
1631 |
+
"epoch": 109.0,
|
1632 |
+
"learning_rate": 1.7542014004668223e-05,
|
1633 |
+
"loss": 2.9834,
|
1634 |
+
"step": 194783
|
1635 |
+
},
|
1636 |
+
{
|
1637 |
+
"epoch": 109.0,
|
1638 |
+
"eval_accuracy": 0.4537605471925546,
|
1639 |
+
"eval_loss": 3.0319700241088867,
|
1640 |
+
"eval_runtime": 3.1336,
|
1641 |
+
"eval_samples_per_second": 152.86,
|
1642 |
+
"eval_steps_per_second": 1.276,
|
1643 |
+
"step": 194783
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 110.0,
|
1647 |
+
"learning_rate": 1.7244081360453486e-05,
|
1648 |
+
"loss": 2.9829,
|
1649 |
+
"step": 196570
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 110.0,
|
1653 |
+
"eval_accuracy": 0.44560168895144264,
|
1654 |
+
"eval_loss": 3.0759708881378174,
|
1655 |
+
"eval_runtime": 2.6847,
|
1656 |
+
"eval_samples_per_second": 178.418,
|
1657 |
+
"eval_steps_per_second": 1.49,
|
1658 |
+
"step": 196570
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 111.0,
|
1662 |
+
"learning_rate": 1.6946148716238746e-05,
|
1663 |
+
"loss": 2.9814,
|
1664 |
+
"step": 198357
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 111.0,
|
1668 |
+
"eval_accuracy": 0.45036670646426746,
|
1669 |
+
"eval_loss": 3.0422720909118652,
|
1670 |
+
"eval_runtime": 2.9828,
|
1671 |
+
"eval_samples_per_second": 160.59,
|
1672 |
+
"eval_steps_per_second": 1.341,
|
1673 |
+
"step": 198357
|
1674 |
+
},
|
1675 |
+
{
|
1676 |
+
"epoch": 112.0,
|
1677 |
+
"learning_rate": 1.664821607202401e-05,
|
1678 |
+
"loss": 2.9795,
|
1679 |
+
"step": 200144
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 112.0,
|
1683 |
+
"eval_accuracy": 0.452860867371254,
|
1684 |
+
"eval_loss": 3.0410661697387695,
|
1685 |
+
"eval_runtime": 2.9476,
|
1686 |
+
"eval_samples_per_second": 162.504,
|
1687 |
+
"eval_steps_per_second": 1.357,
|
1688 |
+
"step": 200144
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 113.0,
|
1692 |
+
"learning_rate": 1.6350283427809272e-05,
|
1693 |
+
"loss": 2.979,
|
1694 |
+
"step": 201931
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 113.0,
|
1698 |
+
"eval_accuracy": 0.4462770034348653,
|
1699 |
+
"eval_loss": 3.0784435272216797,
|
1700 |
+
"eval_runtime": 3.0447,
|
1701 |
+
"eval_samples_per_second": 157.321,
|
1702 |
+
"eval_steps_per_second": 1.314,
|
1703 |
+
"step": 201931
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 114.0,
|
1707 |
+
"learning_rate": 1.6052350783594532e-05,
|
1708 |
+
"loss": 2.9781,
|
1709 |
+
"step": 203718
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 114.0,
|
1713 |
+
"eval_accuracy": 0.453679165491965,
|
1714 |
+
"eval_loss": 3.0525529384613037,
|
1715 |
+
"eval_runtime": 2.985,
|
1716 |
+
"eval_samples_per_second": 160.472,
|
1717 |
+
"eval_steps_per_second": 1.34,
|
1718 |
+
"step": 203718
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 115.0,
|
1722 |
+
"learning_rate": 1.5754418139379795e-05,
|
1723 |
+
"loss": 2.9751,
|
1724 |
+
"step": 205505
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 115.0,
|
1728 |
+
"eval_accuracy": 0.4512479201331115,
|
1729 |
+
"eval_loss": 3.0478885173797607,
|
1730 |
+
"eval_runtime": 3.2329,
|
1731 |
+
"eval_samples_per_second": 148.165,
|
1732 |
+
"eval_steps_per_second": 1.237,
|
1733 |
+
"step": 205505
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 116.0,
|
1737 |
+
"learning_rate": 1.5456485495165055e-05,
|
1738 |
+
"loss": 2.9749,
|
1739 |
+
"step": 207292
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 116.0,
|
1743 |
+
"eval_accuracy": 0.44932773109243695,
|
1744 |
+
"eval_loss": 3.054450511932373,
|
1745 |
+
"eval_runtime": 2.6607,
|
1746 |
+
"eval_samples_per_second": 180.027,
|
1747 |
+
"eval_steps_per_second": 1.503,
|
1748 |
+
"step": 207292
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 117.0,
|
1752 |
+
"learning_rate": 1.5158552850950317e-05,
|
1753 |
+
"loss": 2.9735,
|
1754 |
+
"step": 209079
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 117.0,
|
1758 |
+
"eval_accuracy": 0.44851187971612105,
|
1759 |
+
"eval_loss": 3.0529181957244873,
|
1760 |
+
"eval_runtime": 2.6442,
|
1761 |
+
"eval_samples_per_second": 181.151,
|
1762 |
+
"eval_steps_per_second": 1.513,
|
1763 |
+
"step": 209079
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 118.0,
|
1767 |
+
"learning_rate": 1.4860620206735578e-05,
|
1768 |
+
"loss": 2.9705,
|
1769 |
+
"step": 210866
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 118.0,
|
1773 |
+
"eval_accuracy": 0.4580500769338369,
|
1774 |
+
"eval_loss": 3.008023977279663,
|
1775 |
+
"eval_runtime": 2.8363,
|
1776 |
+
"eval_samples_per_second": 168.883,
|
1777 |
+
"eval_steps_per_second": 1.41,
|
1778 |
+
"step": 210866
|
1779 |
+
},
|
1780 |
+
{
|
1781 |
+
"epoch": 119.0,
|
1782 |
+
"learning_rate": 1.4562687562520841e-05,
|
1783 |
+
"loss": 2.9698,
|
1784 |
+
"step": 212653
|
1785 |
+
},
|
1786 |
+
{
|
1787 |
+
"epoch": 119.0,
|
1788 |
+
"eval_accuracy": 0.4536538892624972,
|
1789 |
+
"eval_loss": 3.0271081924438477,
|
1790 |
+
"eval_runtime": 2.5889,
|
1791 |
+
"eval_samples_per_second": 185.019,
|
1792 |
+
"eval_steps_per_second": 1.545,
|
1793 |
+
"step": 212653
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 120.0,
|
1797 |
+
"learning_rate": 1.4264754918306103e-05,
|
1798 |
+
"loss": 2.9674,
|
1799 |
+
"step": 214440
|
1800 |
+
},
|
1801 |
+
{
|
1802 |
+
"epoch": 120.0,
|
1803 |
+
"eval_accuracy": 0.4481896208719377,
|
1804 |
+
"eval_loss": 3.0476975440979004,
|
1805 |
+
"eval_runtime": 2.6869,
|
1806 |
+
"eval_samples_per_second": 178.272,
|
1807 |
+
"eval_steps_per_second": 1.489,
|
1808 |
+
"step": 214440
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 121.0,
|
1812 |
+
"learning_rate": 1.3966822274091364e-05,
|
1813 |
+
"loss": 2.9666,
|
1814 |
+
"step": 216227
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 121.0,
|
1818 |
+
"eval_accuracy": 0.45575815202192516,
|
1819 |
+
"eval_loss": 3.032849073410034,
|
1820 |
+
"eval_runtime": 2.9433,
|
1821 |
+
"eval_samples_per_second": 162.744,
|
1822 |
+
"eval_steps_per_second": 1.359,
|
1823 |
+
"step": 216227
|
1824 |
+
},
|
1825 |
+
{
|
1826 |
+
"epoch": 122.0,
|
1827 |
+
"learning_rate": 1.3668889629876628e-05,
|
1828 |
+
"loss": 2.9664,
|
1829 |
+
"step": 218014
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 122.0,
|
1833 |
+
"eval_accuracy": 0.4462809917355372,
|
1834 |
+
"eval_loss": 3.068892240524292,
|
1835 |
+
"eval_runtime": 2.5675,
|
1836 |
+
"eval_samples_per_second": 186.559,
|
1837 |
+
"eval_steps_per_second": 1.558,
|
1838 |
+
"step": 218014
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 123.0,
|
1842 |
+
"learning_rate": 1.3370956985661887e-05,
|
1843 |
+
"loss": 2.9639,
|
1844 |
+
"step": 219801
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 123.0,
|
1848 |
+
"eval_accuracy": 0.44586561487533455,
|
1849 |
+
"eval_loss": 3.0748960971832275,
|
1850 |
+
"eval_runtime": 2.796,
|
1851 |
+
"eval_samples_per_second": 171.315,
|
1852 |
+
"eval_steps_per_second": 1.431,
|
1853 |
+
"step": 219801
|
1854 |
+
},
|
1855 |
+
{
|
1856 |
+
"epoch": 124.0,
|
1857 |
+
"learning_rate": 1.3073024341447149e-05,
|
1858 |
+
"loss": 2.9633,
|
1859 |
+
"step": 221588
|
1860 |
+
},
|
1861 |
+
{
|
1862 |
+
"epoch": 124.0,
|
1863 |
+
"eval_accuracy": 0.4488954922628392,
|
1864 |
+
"eval_loss": 3.0504775047302246,
|
1865 |
+
"eval_runtime": 2.7162,
|
1866 |
+
"eval_samples_per_second": 176.347,
|
1867 |
+
"eval_steps_per_second": 1.473,
|
1868 |
+
"step": 221588
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 125.0,
|
1872 |
+
"learning_rate": 1.277509169723241e-05,
|
1873 |
+
"loss": 2.9618,
|
1874 |
+
"step": 223375
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 125.0,
|
1878 |
+
"eval_accuracy": 0.45345192714951293,
|
1879 |
+
"eval_loss": 3.025569438934326,
|
1880 |
+
"eval_runtime": 2.7094,
|
1881 |
+
"eval_samples_per_second": 176.794,
|
1882 |
+
"eval_steps_per_second": 1.476,
|
1883 |
+
"step": 223375
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 126.0,
|
1887 |
+
"learning_rate": 1.2477159053017674e-05,
|
1888 |
+
"loss": 2.9589,
|
1889 |
+
"step": 225162
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 126.0,
|
1893 |
+
"eval_accuracy": 0.4495620785180625,
|
1894 |
+
"eval_loss": 3.052183151245117,
|
1895 |
+
"eval_runtime": 2.6479,
|
1896 |
+
"eval_samples_per_second": 180.901,
|
1897 |
+
"eval_steps_per_second": 1.511,
|
1898 |
+
"step": 225162
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 127.0,
|
1902 |
+
"learning_rate": 1.2179226408802935e-05,
|
1903 |
+
"loss": 2.9584,
|
1904 |
+
"step": 226949
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 127.0,
|
1908 |
+
"eval_accuracy": 0.45301326453349783,
|
1909 |
+
"eval_loss": 3.0450875759124756,
|
1910 |
+
"eval_runtime": 2.6347,
|
1911 |
+
"eval_samples_per_second": 181.807,
|
1912 |
+
"eval_steps_per_second": 1.518,
|
1913 |
+
"step": 226949
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 128.0,
|
1917 |
+
"learning_rate": 1.1881293764588197e-05,
|
1918 |
+
"loss": 2.9589,
|
1919 |
+
"step": 228736
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 128.0,
|
1923 |
+
"eval_accuracy": 0.4501640457065279,
|
1924 |
+
"eval_loss": 3.065441846847534,
|
1925 |
+
"eval_runtime": 2.6108,
|
1926 |
+
"eval_samples_per_second": 183.467,
|
1927 |
+
"eval_steps_per_second": 1.532,
|
1928 |
+
"step": 228736
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 129.0,
|
1932 |
+
"learning_rate": 1.1583361120373458e-05,
|
1933 |
+
"loss": 2.9581,
|
1934 |
+
"step": 230523
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 129.0,
|
1938 |
+
"eval_accuracy": 0.45803146932307,
|
1939 |
+
"eval_loss": 2.998933792114258,
|
1940 |
+
"eval_runtime": 2.5885,
|
1941 |
+
"eval_samples_per_second": 185.049,
|
1942 |
+
"eval_steps_per_second": 1.545,
|
1943 |
+
"step": 230523
|
1944 |
+
},
|
1945 |
+
{
|
1946 |
+
"epoch": 130.0,
|
1947 |
+
"learning_rate": 1.128542847615872e-05,
|
1948 |
+
"loss": 2.9554,
|
1949 |
+
"step": 232310
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 130.0,
|
1953 |
+
"eval_accuracy": 0.45082426825165417,
|
1954 |
+
"eval_loss": 3.0346689224243164,
|
1955 |
+
"eval_runtime": 2.6119,
|
1956 |
+
"eval_samples_per_second": 183.394,
|
1957 |
+
"eval_steps_per_second": 1.531,
|
1958 |
+
"step": 232310
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 131.0,
|
1962 |
+
"learning_rate": 1.0987495831943983e-05,
|
1963 |
+
"loss": 2.9565,
|
1964 |
+
"step": 234097
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 131.0,
|
1968 |
+
"eval_accuracy": 0.44976425684777727,
|
1969 |
+
"eval_loss": 3.0585811138153076,
|
1970 |
+
"eval_runtime": 2.5925,
|
1971 |
+
"eval_samples_per_second": 184.76,
|
1972 |
+
"eval_steps_per_second": 1.543,
|
1973 |
+
"step": 234097
|
1974 |
+
},
|
1975 |
+
{
|
1976 |
+
"epoch": 132.0,
|
1977 |
+
"learning_rate": 1.0689563187729243e-05,
|
1978 |
+
"loss": 2.9548,
|
1979 |
+
"step": 235884
|
1980 |
+
},
|
1981 |
+
{
|
1982 |
+
"epoch": 132.0,
|
1983 |
+
"eval_accuracy": 0.4535522171743917,
|
1984 |
+
"eval_loss": 3.016965389251709,
|
1985 |
+
"eval_runtime": 2.5996,
|
1986 |
+
"eval_samples_per_second": 184.261,
|
1987 |
+
"eval_steps_per_second": 1.539,
|
1988 |
+
"step": 235884
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 133.0,
|
1992 |
+
"learning_rate": 1.0391630543514506e-05,
|
1993 |
+
"loss": 2.9515,
|
1994 |
+
"step": 237671
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 133.0,
|
1998 |
+
"eval_accuracy": 0.44919200987543484,
|
1999 |
+
"eval_loss": 3.046969175338745,
|
2000 |
+
"eval_runtime": 2.7591,
|
2001 |
+
"eval_samples_per_second": 173.609,
|
2002 |
+
"eval_steps_per_second": 1.45,
|
2003 |
+
"step": 237671
|
2004 |
+
},
|
2005 |
+
{
|
2006 |
+
"epoch": 134.0,
|
2007 |
+
"learning_rate": 1.0093697899299767e-05,
|
2008 |
+
"loss": 2.9499,
|
2009 |
+
"step": 239458
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 134.0,
|
2013 |
+
"eval_accuracy": 0.4514692787177204,
|
2014 |
+
"eval_loss": 3.033890724182129,
|
2015 |
+
"eval_runtime": 2.5902,
|
2016 |
+
"eval_samples_per_second": 184.93,
|
2017 |
+
"eval_steps_per_second": 1.544,
|
2018 |
+
"step": 239458
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 135.0,
|
2022 |
+
"learning_rate": 9.795765255085029e-06,
|
2023 |
+
"loss": 2.9514,
|
2024 |
+
"step": 241245
|
2025 |
+
},
|
2026 |
+
{
|
2027 |
+
"epoch": 135.0,
|
2028 |
+
"eval_accuracy": 0.44730392156862747,
|
2029 |
+
"eval_loss": 3.047365665435791,
|
2030 |
+
"eval_runtime": 2.5837,
|
2031 |
+
"eval_samples_per_second": 185.39,
|
2032 |
+
"eval_steps_per_second": 1.548,
|
2033 |
+
"step": 241245
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 136.0,
|
2037 |
+
"learning_rate": 9.49783261087029e-06,
|
2038 |
+
"loss": 2.9486,
|
2039 |
+
"step": 243032
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 136.0,
|
2043 |
+
"eval_accuracy": 0.4492987983383054,
|
2044 |
+
"eval_loss": 3.0426554679870605,
|
2045 |
+
"eval_runtime": 2.6436,
|
2046 |
+
"eval_samples_per_second": 181.191,
|
2047 |
+
"eval_steps_per_second": 1.513,
|
2048 |
+
"step": 243032
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 137.0,
|
2052 |
+
"learning_rate": 9.199899966655552e-06,
|
2053 |
+
"loss": 2.9483,
|
2054 |
+
"step": 244819
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 137.0,
|
2058 |
+
"eval_accuracy": 0.45336896891264594,
|
2059 |
+
"eval_loss": 3.033630847930908,
|
2060 |
+
"eval_runtime": 2.6114,
|
2061 |
+
"eval_samples_per_second": 183.424,
|
2062 |
+
"eval_steps_per_second": 1.532,
|
2063 |
+
"step": 244819
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 138.0,
|
2067 |
+
"learning_rate": 8.901967322440815e-06,
|
2068 |
+
"loss": 2.9491,
|
2069 |
+
"step": 246606
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 138.0,
|
2073 |
+
"eval_accuracy": 0.4516002576092739,
|
2074 |
+
"eval_loss": 3.027387857437134,
|
2075 |
+
"eval_runtime": 2.5975,
|
2076 |
+
"eval_samples_per_second": 184.405,
|
2077 |
+
"eval_steps_per_second": 1.54,
|
2078 |
+
"step": 246606
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 139.0,
|
2082 |
+
"learning_rate": 8.604034678226076e-06,
|
2083 |
+
"loss": 2.9465,
|
2084 |
+
"step": 248393
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 139.0,
|
2088 |
+
"eval_accuracy": 0.45385626765882453,
|
2089 |
+
"eval_loss": 3.0354230403900146,
|
2090 |
+
"eval_runtime": 2.5982,
|
2091 |
+
"eval_samples_per_second": 184.359,
|
2092 |
+
"eval_steps_per_second": 1.54,
|
2093 |
+
"step": 248393
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 140.0,
|
2097 |
+
"learning_rate": 8.306102034011338e-06,
|
2098 |
+
"loss": 2.9447,
|
2099 |
+
"step": 250180
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 140.0,
|
2103 |
+
"eval_accuracy": 0.4525785166095551,
|
2104 |
+
"eval_loss": 3.013915538787842,
|
2105 |
+
"eval_runtime": 2.7656,
|
2106 |
+
"eval_samples_per_second": 173.197,
|
2107 |
+
"eval_steps_per_second": 1.446,
|
2108 |
+
"step": 250180
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 141.0,
|
2112 |
+
"learning_rate": 8.0081693897966e-06,
|
2113 |
+
"loss": 2.9449,
|
2114 |
+
"step": 251967
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 141.0,
|
2118 |
+
"eval_accuracy": 0.4547931354099374,
|
2119 |
+
"eval_loss": 3.016339063644409,
|
2120 |
+
"eval_runtime": 2.7347,
|
2121 |
+
"eval_samples_per_second": 175.155,
|
2122 |
+
"eval_steps_per_second": 1.463,
|
2123 |
+
"step": 251967
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 142.0,
|
2127 |
+
"learning_rate": 7.710236745581861e-06,
|
2128 |
+
"loss": 2.9439,
|
2129 |
+
"step": 253754
|
2130 |
+
},
|
2131 |
+
{
|
2132 |
+
"epoch": 142.0,
|
2133 |
+
"eval_accuracy": 0.45336875423920414,
|
2134 |
+
"eval_loss": 3.0307540893554688,
|
2135 |
+
"eval_runtime": 2.688,
|
2136 |
+
"eval_samples_per_second": 178.198,
|
2137 |
+
"eval_steps_per_second": 1.488,
|
2138 |
+
"step": 253754
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 143.0,
|
2142 |
+
"learning_rate": 7.412304101367122e-06,
|
2143 |
+
"loss": 2.9435,
|
2144 |
+
"step": 255541
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 143.0,
|
2148 |
+
"eval_accuracy": 0.4578975260089811,
|
2149 |
+
"eval_loss": 3.0242276191711426,
|
2150 |
+
"eval_runtime": 2.663,
|
2151 |
+
"eval_samples_per_second": 179.874,
|
2152 |
+
"eval_steps_per_second": 1.502,
|
2153 |
+
"step": 255541
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 144.0,
|
2157 |
+
"learning_rate": 7.114371457152384e-06,
|
2158 |
+
"loss": 2.943,
|
2159 |
+
"step": 257328
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 144.0,
|
2163 |
+
"eval_accuracy": 0.45133795837462837,
|
2164 |
+
"eval_loss": 3.043705463409424,
|
2165 |
+
"eval_runtime": 2.7686,
|
2166 |
+
"eval_samples_per_second": 173.01,
|
2167 |
+
"eval_steps_per_second": 1.445,
|
2168 |
+
"step": 257328
|
2169 |
+
},
|
2170 |
+
{
|
2171 |
+
"epoch": 145.0,
|
2172 |
+
"learning_rate": 6.816438812937646e-06,
|
2173 |
+
"loss": 2.943,
|
2174 |
+
"step": 259115
|
2175 |
+
},
|
2176 |
+
{
|
2177 |
+
"epoch": 145.0,
|
2178 |
+
"eval_accuracy": 0.4544336352454434,
|
2179 |
+
"eval_loss": 3.022679328918457,
|
2180 |
+
"eval_runtime": 2.8352,
|
2181 |
+
"eval_samples_per_second": 168.947,
|
2182 |
+
"eval_steps_per_second": 1.411,
|
2183 |
+
"step": 259115
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 146.0,
|
2187 |
+
"learning_rate": 6.518506168722909e-06,
|
2188 |
+
"loss": 2.9403,
|
2189 |
+
"step": 260902
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 146.0,
|
2193 |
+
"eval_accuracy": 0.4477586597237944,
|
2194 |
+
"eval_loss": 3.0463666915893555,
|
2195 |
+
"eval_runtime": 2.8089,
|
2196 |
+
"eval_samples_per_second": 170.531,
|
2197 |
+
"eval_steps_per_second": 1.424,
|
2198 |
+
"step": 260902
|
2199 |
+
},
|
2200 |
+
{
|
2201 |
+
"epoch": 147.0,
|
2202 |
+
"learning_rate": 6.22057352450817e-06,
|
2203 |
+
"loss": 2.9407,
|
2204 |
+
"step": 262689
|
2205 |
+
},
|
2206 |
+
{
|
2207 |
+
"epoch": 147.0,
|
2208 |
+
"eval_accuracy": 0.4465163301412564,
|
2209 |
+
"eval_loss": 3.0717713832855225,
|
2210 |
+
"eval_runtime": 2.6924,
|
2211 |
+
"eval_samples_per_second": 177.907,
|
2212 |
+
"eval_steps_per_second": 1.486,
|
2213 |
+
"step": 262689
|
2214 |
+
},
|
2215 |
+
{
|
2216 |
+
"epoch": 148.0,
|
2217 |
+
"learning_rate": 5.922640880293432e-06,
|
2218 |
+
"loss": 2.9397,
|
2219 |
+
"step": 264476
|
2220 |
+
},
|
2221 |
+
{
|
2222 |
+
"epoch": 148.0,
|
2223 |
+
"eval_accuracy": 0.4487103426844083,
|
2224 |
+
"eval_loss": 3.0518863201141357,
|
2225 |
+
"eval_runtime": 2.6462,
|
2226 |
+
"eval_samples_per_second": 181.012,
|
2227 |
+
"eval_steps_per_second": 1.512,
|
2228 |
+
"step": 264476
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 149.0,
|
2232 |
+
"learning_rate": 5.624708236078693e-06,
|
2233 |
+
"loss": 2.9392,
|
2234 |
+
"step": 266263
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 149.0,
|
2238 |
+
"eval_accuracy": 0.45580133672641854,
|
2239 |
+
"eval_loss": 3.0162956714630127,
|
2240 |
+
"eval_runtime": 2.8146,
|
2241 |
+
"eval_samples_per_second": 170.186,
|
2242 |
+
"eval_steps_per_second": 1.421,
|
2243 |
+
"step": 266263
|
2244 |
+
},
|
2245 |
+
{
|
2246 |
+
"epoch": 150.0,
|
2247 |
+
"learning_rate": 5.326775591863955e-06,
|
2248 |
+
"loss": 2.9377,
|
2249 |
+
"step": 268050
|
2250 |
+
},
|
2251 |
+
{
|
2252 |
+
"epoch": 150.0,
|
2253 |
+
"eval_accuracy": 0.45181576114667726,
|
2254 |
+
"eval_loss": 3.015916585922241,
|
2255 |
+
"eval_runtime": 2.7189,
|
2256 |
+
"eval_samples_per_second": 176.176,
|
2257 |
+
"eval_steps_per_second": 1.471,
|
2258 |
+
"step": 268050
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 151.0,
|
2262 |
+
"learning_rate": 5.028842947649216e-06,
|
2263 |
+
"loss": 2.9386,
|
2264 |
+
"step": 269837
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 151.0,
|
2268 |
+
"eval_accuracy": 0.45453775343084657,
|
2269 |
+
"eval_loss": 3.0009806156158447,
|
2270 |
+
"eval_runtime": 2.8171,
|
2271 |
+
"eval_samples_per_second": 170.031,
|
2272 |
+
"eval_steps_per_second": 1.42,
|
2273 |
+
"step": 269837
|
2274 |
+
},
|
2275 |
+
{
|
2276 |
+
"epoch": 152.0,
|
2277 |
+
"learning_rate": 4.7309103034344784e-06,
|
2278 |
+
"loss": 2.9391,
|
2279 |
+
"step": 271624
|
2280 |
+
},
|
2281 |
+
{
|
2282 |
+
"epoch": 152.0,
|
2283 |
+
"eval_accuracy": 0.4530369705973036,
|
2284 |
+
"eval_loss": 3.034574270248413,
|
2285 |
+
"eval_runtime": 2.7339,
|
2286 |
+
"eval_samples_per_second": 175.207,
|
2287 |
+
"eval_steps_per_second": 1.463,
|
2288 |
+
"step": 271624
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 153.0,
|
2292 |
+
"learning_rate": 4.43297765921974e-06,
|
2293 |
+
"loss": 2.9364,
|
2294 |
+
"step": 273411
|
2295 |
+
},
|
2296 |
+
{
|
2297 |
+
"epoch": 153.0,
|
2298 |
+
"eval_accuracy": 0.45406984833727565,
|
2299 |
+
"eval_loss": 3.00394606590271,
|
2300 |
+
"eval_runtime": 2.6952,
|
2301 |
+
"eval_samples_per_second": 177.722,
|
2302 |
+
"eval_steps_per_second": 1.484,
|
2303 |
+
"step": 273411
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 154.0,
|
2307 |
+
"learning_rate": 4.135045015005002e-06,
|
2308 |
+
"loss": 2.9359,
|
2309 |
+
"step": 275198
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 154.0,
|
2313 |
+
"eval_accuracy": 0.4519481963203824,
|
2314 |
+
"eval_loss": 3.041689157485962,
|
2315 |
+
"eval_runtime": 2.7367,
|
2316 |
+
"eval_samples_per_second": 175.028,
|
2317 |
+
"eval_steps_per_second": 1.462,
|
2318 |
+
"step": 275198
|
2319 |
+
},
|
2320 |
+
{
|
2321 |
+
"epoch": 155.0,
|
2322 |
+
"learning_rate": 3.837112370790264e-06,
|
2323 |
+
"loss": 2.9359,
|
2324 |
+
"step": 276985
|
2325 |
+
},
|
2326 |
+
{
|
2327 |
+
"epoch": 155.0,
|
2328 |
+
"eval_accuracy": 0.45435859894132224,
|
2329 |
+
"eval_loss": 3.0161099433898926,
|
2330 |
+
"eval_runtime": 2.74,
|
2331 |
+
"eval_samples_per_second": 174.817,
|
2332 |
+
"eval_steps_per_second": 1.46,
|
2333 |
+
"step": 276985
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 156.0,
|
2337 |
+
"learning_rate": 3.5391797265755257e-06,
|
2338 |
+
"loss": 2.936,
|
2339 |
+
"step": 278772
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 156.0,
|
2343 |
+
"eval_accuracy": 0.45336758628392715,
|
2344 |
+
"eval_loss": 3.0168893337249756,
|
2345 |
+
"eval_runtime": 2.8607,
|
2346 |
+
"eval_samples_per_second": 167.443,
|
2347 |
+
"eval_steps_per_second": 1.398,
|
2348 |
+
"step": 278772
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 157.0,
|
2352 |
+
"learning_rate": 3.241247082360787e-06,
|
2353 |
+
"loss": 2.9329,
|
2354 |
+
"step": 280559
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 157.0,
|
2358 |
+
"eval_accuracy": 0.4478249866434215,
|
2359 |
+
"eval_loss": 3.059436082839966,
|
2360 |
+
"eval_runtime": 2.856,
|
2361 |
+
"eval_samples_per_second": 167.714,
|
2362 |
+
"eval_steps_per_second": 1.401,
|
2363 |
+
"step": 280559
|
2364 |
+
},
|
2365 |
+
{
|
2366 |
+
"epoch": 158.0,
|
2367 |
+
"learning_rate": 2.9433144381460487e-06,
|
2368 |
+
"loss": 2.9336,
|
2369 |
+
"step": 282346
|
2370 |
+
},
|
2371 |
+
{
|
2372 |
+
"epoch": 158.0,
|
2373 |
+
"eval_accuracy": 0.45551214066424783,
|
2374 |
+
"eval_loss": 3.0264980792999268,
|
2375 |
+
"eval_runtime": 2.9513,
|
2376 |
+
"eval_samples_per_second": 162.304,
|
2377 |
+
"eval_steps_per_second": 1.355,
|
2378 |
+
"step": 282346
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 159.0,
|
2382 |
+
"learning_rate": 2.6453817939313106e-06,
|
2383 |
+
"loss": 2.9341,
|
2384 |
+
"step": 284133
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 159.0,
|
2388 |
+
"eval_accuracy": 0.4542034031560622,
|
2389 |
+
"eval_loss": 3.0276126861572266,
|
2390 |
+
"eval_runtime": 2.8692,
|
2391 |
+
"eval_samples_per_second": 166.946,
|
2392 |
+
"eval_steps_per_second": 1.394,
|
2393 |
+
"step": 284133
|
2394 |
+
},
|
2395 |
+
{
|
2396 |
+
"epoch": 160.0,
|
2397 |
+
"learning_rate": 2.347449149716572e-06,
|
2398 |
+
"loss": 2.933,
|
2399 |
+
"step": 285920
|
2400 |
+
},
|
2401 |
+
{
|
2402 |
+
"epoch": 160.0,
|
2403 |
+
"eval_accuracy": 0.4524090960261173,
|
2404 |
+
"eval_loss": 3.032360553741455,
|
2405 |
+
"eval_runtime": 2.7445,
|
2406 |
+
"eval_samples_per_second": 174.528,
|
2407 |
+
"eval_steps_per_second": 1.457,
|
2408 |
+
"step": 285920
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 161.0,
|
2412 |
+
"learning_rate": 2.049516505501834e-06,
|
2413 |
+
"loss": 2.9325,
|
2414 |
+
"step": 287707
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 161.0,
|
2418 |
+
"eval_accuracy": 0.4489003880983182,
|
2419 |
+
"eval_loss": 3.024897813796997,
|
2420 |
+
"eval_runtime": 2.949,
|
2421 |
+
"eval_samples_per_second": 162.427,
|
2422 |
+
"eval_steps_per_second": 1.356,
|
2423 |
+
"step": 287707
|
2424 |
+
},
|
2425 |
+
{
|
2426 |
+
"epoch": 162.0,
|
2427 |
+
"learning_rate": 1.7515838612870958e-06,
|
2428 |
+
"loss": 2.932,
|
2429 |
+
"step": 289494
|
2430 |
+
},
|
2431 |
+
{
|
2432 |
+
"epoch": 162.0,
|
2433 |
+
"eval_accuracy": 0.4518854201633658,
|
2434 |
+
"eval_loss": 3.0444188117980957,
|
2435 |
+
"eval_runtime": 3.2716,
|
2436 |
+
"eval_samples_per_second": 146.412,
|
2437 |
+
"eval_steps_per_second": 1.223,
|
2438 |
+
"step": 289494
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 163.0,
|
2442 |
+
"learning_rate": 1.4536512170723575e-06,
|
2443 |
+
"loss": 2.9334,
|
2444 |
+
"step": 291281
|
2445 |
+
},
|
2446 |
+
{
|
2447 |
+
"epoch": 163.0,
|
2448 |
+
"eval_accuracy": 0.4493931664657903,
|
2449 |
+
"eval_loss": 3.0419514179229736,
|
2450 |
+
"eval_runtime": 3.072,
|
2451 |
+
"eval_samples_per_second": 155.926,
|
2452 |
+
"eval_steps_per_second": 1.302,
|
2453 |
+
"step": 291281
|
2454 |
+
},
|
2455 |
+
{
|
2456 |
+
"epoch": 164.0,
|
2457 |
+
"learning_rate": 1.1557185728576192e-06,
|
2458 |
+
"loss": 2.9318,
|
2459 |
+
"step": 293068
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 164.0,
|
2463 |
+
"eval_accuracy": 0.454113304625021,
|
2464 |
+
"eval_loss": 2.9971697330474854,
|
2465 |
+
"eval_runtime": 3.0501,
|
2466 |
+
"eval_samples_per_second": 157.044,
|
2467 |
+
"eval_steps_per_second": 1.311,
|
2468 |
+
"step": 293068
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 165.0,
|
2472 |
+
"learning_rate": 8.57785928642881e-07,
|
2473 |
+
"loss": 2.9316,
|
2474 |
+
"step": 294855
|
2475 |
+
},
|
2476 |
+
{
|
2477 |
+
"epoch": 165.0,
|
2478 |
+
"eval_accuracy": 0.4525653024759636,
|
2479 |
+
"eval_loss": 2.997295618057251,
|
2480 |
+
"eval_runtime": 2.7549,
|
2481 |
+
"eval_samples_per_second": 173.872,
|
2482 |
+
"eval_steps_per_second": 1.452,
|
2483 |
+
"step": 294855
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 166.0,
|
2487 |
+
"learning_rate": 5.598532844281427e-07,
|
2488 |
+
"loss": 2.9318,
|
2489 |
+
"step": 296642
|
2490 |
+
},
|
2491 |
+
{
|
2492 |
+
"epoch": 166.0,
|
2493 |
+
"eval_accuracy": 0.45290194408968354,
|
2494 |
+
"eval_loss": 3.0388541221618652,
|
2495 |
+
"eval_runtime": 3.0075,
|
2496 |
+
"eval_samples_per_second": 159.266,
|
2497 |
+
"eval_steps_per_second": 1.33,
|
2498 |
+
"step": 296642
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 167.0,
|
2502 |
+
"learning_rate": 2.619206402134045e-07,
|
2503 |
+
"loss": 2.9301,
|
2504 |
+
"step": 298429
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 167.0,
|
2508 |
+
"eval_accuracy": 0.4556901408450704,
|
2509 |
+
"eval_loss": 3.013132095336914,
|
2510 |
+
"eval_runtime": 2.9991,
|
2511 |
+
"eval_samples_per_second": 159.713,
|
2512 |
+
"eval_steps_per_second": 1.334,
|
2513 |
+
"step": 298429
|
2514 |
+
},
|
2515 |
+
{
|
2516 |
+
"epoch": 167.88,
|
2517 |
+
"learning_rate": 0.0,
|
2518 |
+
"loss": 2.9291,
|
2519 |
+
"step": 300000
|
2520 |
+
},
|
2521 |
+
{
|
2522 |
+
"epoch": 167.88,
|
2523 |
+
"eval_accuracy": 0.4548123706727812,
|
2524 |
+
"eval_loss": 3.006730794906616,
|
2525 |
+
"eval_runtime": 3.1035,
|
2526 |
+
"eval_samples_per_second": 154.341,
|
2527 |
+
"eval_steps_per_second": 1.289,
|
2528 |
+
"step": 300000
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 167.88,
|
2532 |
+
"step": 300000,
|
2533 |
+
"total_flos": 3.687416469441741e+18,
|
2534 |
+
"train_loss": 3.1306004736328124,
|
2535 |
+
"train_runtime": 219827.3924,
|
2536 |
+
"train_samples_per_second": 174.683,
|
2537 |
+
"train_steps_per_second": 1.365
|
2538 |
+
}
|
2539 |
+
],
|
2540 |
+
"max_steps": 300000,
|
2541 |
+
"num_train_epochs": 168,
|
2542 |
+
"total_flos": 3.687416469441741e+18,
|
2543 |
+
"trial_name": null,
|
2544 |
+
"trial_params": null
|
2545 |
+
}
|