Training in progress, step 448, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 891644712
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59dd4181cb7b12c18778fce401b831547648cbbc21eaca760b5c6c36b98a4637
|
| 3 |
size 891644712
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1783444794
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d37828ba5932074afc8a3ca7276b64a27b5b48490193b1197d05ecc97f5a3313
|
| 3 |
size 1783444794
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ba0efe6975e5f5bc0ac8e3f88ad39f73f4c39d66d2d03cc9bdba96c5706b8d3
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7fc32074b9fff6edfb63f80aa132a15f80c948014a8e656aa4df28416ac2e074
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -42,13 +42,1546 @@
|
|
| 42 |
"learning_rate": 1.834862385321101e-05,
|
| 43 |
"loss": 12.5906,
|
| 44 |
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
}
|
| 46 |
],
|
| 47 |
"logging_steps": 2,
|
| 48 |
"max_steps": 3606,
|
| 49 |
"num_input_tokens_seen": 0,
|
| 50 |
"num_train_epochs": 2,
|
| 51 |
-
"save_steps":
|
| 52 |
"stateful_callbacks": {
|
| 53 |
"TrainerControl": {
|
| 54 |
"args": {
|
|
@@ -61,7 +1594,7 @@
|
|
| 61 |
"attributes": {}
|
| 62 |
}
|
| 63 |
},
|
| 64 |
-
"total_flos":
|
| 65 |
"train_batch_size": 8,
|
| 66 |
"trial_name": null,
|
| 67 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.24847476428175264,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 448,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 42 |
"learning_rate": 1.834862385321101e-05,
|
| 43 |
"loss": 12.5906,
|
| 44 |
"step": 10
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0066555740432612314,
|
| 48 |
+
"grad_norm": 44.9824333190918,
|
| 49 |
+
"learning_rate": 2.2018348623853213e-05,
|
| 50 |
+
"loss": 11.0436,
|
| 51 |
+
"step": 12
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.00776483638380477,
|
| 55 |
+
"grad_norm": 34.28094482421875,
|
| 56 |
+
"learning_rate": 2.5688073394495416e-05,
|
| 57 |
+
"loss": 9.9942,
|
| 58 |
+
"step": 14
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.00887409872434831,
|
| 62 |
+
"grad_norm": 28.417001724243164,
|
| 63 |
+
"learning_rate": 2.9357798165137618e-05,
|
| 64 |
+
"loss": 9.6041,
|
| 65 |
+
"step": 16
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.009983361064891847,
|
| 69 |
+
"grad_norm": 26.446611404418945,
|
| 70 |
+
"learning_rate": 3.302752293577982e-05,
|
| 71 |
+
"loss": 8.4555,
|
| 72 |
+
"step": 18
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.011092623405435386,
|
| 76 |
+
"grad_norm": 29.56342315673828,
|
| 77 |
+
"learning_rate": 3.669724770642202e-05,
|
| 78 |
+
"loss": 7.7889,
|
| 79 |
+
"step": 20
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.012201885745978924,
|
| 83 |
+
"grad_norm": 62.20895004272461,
|
| 84 |
+
"learning_rate": 4.036697247706422e-05,
|
| 85 |
+
"loss": 6.0004,
|
| 86 |
+
"step": 22
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.013311148086522463,
|
| 90 |
+
"grad_norm": 26.70697784423828,
|
| 91 |
+
"learning_rate": 4.403669724770643e-05,
|
| 92 |
+
"loss": 5.2102,
|
| 93 |
+
"step": 24
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.014420410427066,
|
| 97 |
+
"grad_norm": 20.650150299072266,
|
| 98 |
+
"learning_rate": 4.7706422018348626e-05,
|
| 99 |
+
"loss": 3.786,
|
| 100 |
+
"step": 26
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.01552967276760954,
|
| 104 |
+
"grad_norm": 18.51991081237793,
|
| 105 |
+
"learning_rate": 5.137614678899083e-05,
|
| 106 |
+
"loss": 2.7969,
|
| 107 |
+
"step": 28
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.016638935108153077,
|
| 111 |
+
"grad_norm": 11.624157905578613,
|
| 112 |
+
"learning_rate": 5.504587155963303e-05,
|
| 113 |
+
"loss": 1.9075,
|
| 114 |
+
"step": 30
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.01774819744869662,
|
| 118 |
+
"grad_norm": 4.947307109832764,
|
| 119 |
+
"learning_rate": 5.8715596330275236e-05,
|
| 120 |
+
"loss": 1.4263,
|
| 121 |
+
"step": 32
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.018857459789240156,
|
| 125 |
+
"grad_norm": 2.2392783164978027,
|
| 126 |
+
"learning_rate": 6.238532110091744e-05,
|
| 127 |
+
"loss": 1.3292,
|
| 128 |
+
"step": 34
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.019966722129783693,
|
| 132 |
+
"grad_norm": 1.7417014837265015,
|
| 133 |
+
"learning_rate": 6.605504587155963e-05,
|
| 134 |
+
"loss": 1.128,
|
| 135 |
+
"step": 36
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.02107598447032723,
|
| 139 |
+
"grad_norm": 1.061747431755066,
|
| 140 |
+
"learning_rate": 6.972477064220184e-05,
|
| 141 |
+
"loss": 0.9133,
|
| 142 |
+
"step": 38
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.022185246810870772,
|
| 146 |
+
"grad_norm": 1.0621497631072998,
|
| 147 |
+
"learning_rate": 7.339449541284404e-05,
|
| 148 |
+
"loss": 0.8753,
|
| 149 |
+
"step": 40
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.02329450915141431,
|
| 153 |
+
"grad_norm": 0.772523045539856,
|
| 154 |
+
"learning_rate": 7.706422018348625e-05,
|
| 155 |
+
"loss": 0.6989,
|
| 156 |
+
"step": 42
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.024403771491957847,
|
| 160 |
+
"grad_norm": 0.6044029593467712,
|
| 161 |
+
"learning_rate": 8.073394495412844e-05,
|
| 162 |
+
"loss": 0.7296,
|
| 163 |
+
"step": 44
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.025513033832501388,
|
| 167 |
+
"grad_norm": 0.5906920433044434,
|
| 168 |
+
"learning_rate": 8.440366972477065e-05,
|
| 169 |
+
"loss": 0.6963,
|
| 170 |
+
"step": 46
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.026622296173044926,
|
| 174 |
+
"grad_norm": 0.5078668594360352,
|
| 175 |
+
"learning_rate": 8.807339449541285e-05,
|
| 176 |
+
"loss": 0.6978,
|
| 177 |
+
"step": 48
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.027731558513588463,
|
| 181 |
+
"grad_norm": 0.43668779730796814,
|
| 182 |
+
"learning_rate": 9.174311926605506e-05,
|
| 183 |
+
"loss": 0.6444,
|
| 184 |
+
"step": 50
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.028840820854132,
|
| 188 |
+
"grad_norm": 0.5769929885864258,
|
| 189 |
+
"learning_rate": 9.541284403669725e-05,
|
| 190 |
+
"loss": 0.6321,
|
| 191 |
+
"step": 52
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.029950083194675542,
|
| 195 |
+
"grad_norm": 0.6032134890556335,
|
| 196 |
+
"learning_rate": 9.908256880733946e-05,
|
| 197 |
+
"loss": 0.6399,
|
| 198 |
+
"step": 54
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.03105934553521908,
|
| 202 |
+
"grad_norm": 0.5968932509422302,
|
| 203 |
+
"learning_rate": 0.00010275229357798166,
|
| 204 |
+
"loss": 0.6707,
|
| 205 |
+
"step": 56
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.03216860787576262,
|
| 209 |
+
"grad_norm": 0.5290119051933289,
|
| 210 |
+
"learning_rate": 0.00010642201834862387,
|
| 211 |
+
"loss": 0.8516,
|
| 212 |
+
"step": 58
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.033277870216306155,
|
| 216 |
+
"grad_norm": 0.4980998635292053,
|
| 217 |
+
"learning_rate": 0.00011009174311926606,
|
| 218 |
+
"loss": 0.6211,
|
| 219 |
+
"step": 60
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.03438713255684969,
|
| 223 |
+
"grad_norm": 0.6519685387611389,
|
| 224 |
+
"learning_rate": 0.00011376146788990827,
|
| 225 |
+
"loss": 0.7251,
|
| 226 |
+
"step": 62
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.03549639489739324,
|
| 230 |
+
"grad_norm": 0.6212429404258728,
|
| 231 |
+
"learning_rate": 0.00011743119266055047,
|
| 232 |
+
"loss": 0.4878,
|
| 233 |
+
"step": 64
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.036605657237936774,
|
| 237 |
+
"grad_norm": 0.7839425802230835,
|
| 238 |
+
"learning_rate": 0.00012110091743119268,
|
| 239 |
+
"loss": 0.6157,
|
| 240 |
+
"step": 66
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.03771491957848031,
|
| 244 |
+
"grad_norm": 0.5585289597511292,
|
| 245 |
+
"learning_rate": 0.00012477064220183488,
|
| 246 |
+
"loss": 0.6334,
|
| 247 |
+
"step": 68
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.03882418191902385,
|
| 251 |
+
"grad_norm": 0.4180700480937958,
|
| 252 |
+
"learning_rate": 0.00012844036697247707,
|
| 253 |
+
"loss": 0.639,
|
| 254 |
+
"step": 70
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.03993344425956739,
|
| 258 |
+
"grad_norm": 0.4288474917411804,
|
| 259 |
+
"learning_rate": 0.00013211009174311927,
|
| 260 |
+
"loss": 0.6489,
|
| 261 |
+
"step": 72
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.041042706600110924,
|
| 265 |
+
"grad_norm": 0.391681045293808,
|
| 266 |
+
"learning_rate": 0.00013577981651376149,
|
| 267 |
+
"loss": 0.53,
|
| 268 |
+
"step": 74
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.04215196894065446,
|
| 272 |
+
"grad_norm": 0.4582154154777527,
|
| 273 |
+
"learning_rate": 0.00013944954128440368,
|
| 274 |
+
"loss": 0.6368,
|
| 275 |
+
"step": 76
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.04326123128119801,
|
| 279 |
+
"grad_norm": 0.5818586945533752,
|
| 280 |
+
"learning_rate": 0.0001431192660550459,
|
| 281 |
+
"loss": 0.5243,
|
| 282 |
+
"step": 78
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.044370493621741544,
|
| 286 |
+
"grad_norm": 0.5001092553138733,
|
| 287 |
+
"learning_rate": 0.0001467889908256881,
|
| 288 |
+
"loss": 0.4512,
|
| 289 |
+
"step": 80
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.04547975596228508,
|
| 293 |
+
"grad_norm": 0.37270447611808777,
|
| 294 |
+
"learning_rate": 0.00015045871559633028,
|
| 295 |
+
"loss": 0.582,
|
| 296 |
+
"step": 82
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.04658901830282862,
|
| 300 |
+
"grad_norm": 0.3746323585510254,
|
| 301 |
+
"learning_rate": 0.0001541284403669725,
|
| 302 |
+
"loss": 0.4406,
|
| 303 |
+
"step": 84
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.04769828064337216,
|
| 307 |
+
"grad_norm": 0.3230627775192261,
|
| 308 |
+
"learning_rate": 0.0001577981651376147,
|
| 309 |
+
"loss": 0.4755,
|
| 310 |
+
"step": 86
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.048807542983915694,
|
| 314 |
+
"grad_norm": 0.46364355087280273,
|
| 315 |
+
"learning_rate": 0.00016146788990825688,
|
| 316 |
+
"loss": 0.7201,
|
| 317 |
+
"step": 88
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.04991680532445923,
|
| 321 |
+
"grad_norm": 0.37719470262527466,
|
| 322 |
+
"learning_rate": 0.0001651376146788991,
|
| 323 |
+
"loss": 0.5318,
|
| 324 |
+
"step": 90
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.051026067665002776,
|
| 328 |
+
"grad_norm": 0.3767435550689697,
|
| 329 |
+
"learning_rate": 0.0001688073394495413,
|
| 330 |
+
"loss": 0.4638,
|
| 331 |
+
"step": 92
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.052135330005546314,
|
| 335 |
+
"grad_norm": 0.42447295784950256,
|
| 336 |
+
"learning_rate": 0.00017247706422018351,
|
| 337 |
+
"loss": 0.6399,
|
| 338 |
+
"step": 94
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.05324459234608985,
|
| 342 |
+
"grad_norm": 0.33117732405662537,
|
| 343 |
+
"learning_rate": 0.0001761467889908257,
|
| 344 |
+
"loss": 0.4514,
|
| 345 |
+
"step": 96
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.05435385468663339,
|
| 349 |
+
"grad_norm": 0.32511624693870544,
|
| 350 |
+
"learning_rate": 0.0001798165137614679,
|
| 351 |
+
"loss": 0.5366,
|
| 352 |
+
"step": 98
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.05546311702717693,
|
| 356 |
+
"grad_norm": 0.34356439113616943,
|
| 357 |
+
"learning_rate": 0.00018348623853211012,
|
| 358 |
+
"loss": 0.6371,
|
| 359 |
+
"step": 100
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.056572379367720464,
|
| 363 |
+
"grad_norm": 0.4009954035282135,
|
| 364 |
+
"learning_rate": 0.0001871559633027523,
|
| 365 |
+
"loss": 0.5844,
|
| 366 |
+
"step": 102
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.057681641708264,
|
| 370 |
+
"grad_norm": 0.2896111309528351,
|
| 371 |
+
"learning_rate": 0.0001908256880733945,
|
| 372 |
+
"loss": 0.4478,
|
| 373 |
+
"step": 104
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.058790904048807546,
|
| 377 |
+
"grad_norm": 0.31993618607521057,
|
| 378 |
+
"learning_rate": 0.00019449541284403672,
|
| 379 |
+
"loss": 0.3886,
|
| 380 |
+
"step": 106
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.059900166389351084,
|
| 384 |
+
"grad_norm": 0.44810792803764343,
|
| 385 |
+
"learning_rate": 0.0001981651376146789,
|
| 386 |
+
"loss": 0.5448,
|
| 387 |
+
"step": 108
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.06100942872989462,
|
| 391 |
+
"grad_norm": 0.31323954463005066,
|
| 392 |
+
"learning_rate": 0.00019999995964675577,
|
| 393 |
+
"loss": 0.5082,
|
| 394 |
+
"step": 110
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.06211869107043816,
|
| 398 |
+
"grad_norm": 0.3845660090446472,
|
| 399 |
+
"learning_rate": 0.00019999963682099735,
|
| 400 |
+
"loss": 0.52,
|
| 401 |
+
"step": 112
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.0632279534109817,
|
| 405 |
+
"grad_norm": 0.41813963651657104,
|
| 406 |
+
"learning_rate": 0.00019999899117052264,
|
| 407 |
+
"loss": 0.6146,
|
| 408 |
+
"step": 114
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.06433721575152523,
|
| 412 |
+
"grad_norm": 0.35359835624694824,
|
| 413 |
+
"learning_rate": 0.000199998022697416,
|
| 414 |
+
"loss": 0.5561,
|
| 415 |
+
"step": 116
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.06544647809206877,
|
| 419 |
+
"grad_norm": 0.3362332284450531,
|
| 420 |
+
"learning_rate": 0.0001999967314048039,
|
| 421 |
+
"loss": 0.4836,
|
| 422 |
+
"step": 118
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.06655574043261231,
|
| 426 |
+
"grad_norm": 0.5304926633834839,
|
| 427 |
+
"learning_rate": 0.000199995117296855,
|
| 428 |
+
"loss": 0.5111,
|
| 429 |
+
"step": 120
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.06766500277315585,
|
| 433 |
+
"grad_norm": 0.2606920599937439,
|
| 434 |
+
"learning_rate": 0.00019999318037877995,
|
| 435 |
+
"loss": 0.5379,
|
| 436 |
+
"step": 122
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.06877426511369938,
|
| 440 |
+
"grad_norm": 0.47892895340919495,
|
| 441 |
+
"learning_rate": 0.0001999909206568318,
|
| 442 |
+
"loss": 0.5963,
|
| 443 |
+
"step": 124
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.06988352745424292,
|
| 447 |
+
"grad_norm": 0.30522894859313965,
|
| 448 |
+
"learning_rate": 0.00019998833813830534,
|
| 449 |
+
"loss": 0.6176,
|
| 450 |
+
"step": 126
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.07099278979478647,
|
| 454 |
+
"grad_norm": 0.41433125734329224,
|
| 455 |
+
"learning_rate": 0.00019998543283153772,
|
| 456 |
+
"loss": 0.5134,
|
| 457 |
+
"step": 128
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.07210205213533001,
|
| 461 |
+
"grad_norm": 0.27291831374168396,
|
| 462 |
+
"learning_rate": 0.000199982204745908,
|
| 463 |
+
"loss": 0.5248,
|
| 464 |
+
"step": 130
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.07321131447587355,
|
| 468 |
+
"grad_norm": 0.26184335350990295,
|
| 469 |
+
"learning_rate": 0.0001999786538918372,
|
| 470 |
+
"loss": 0.4302,
|
| 471 |
+
"step": 132
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.07432057681641709,
|
| 475 |
+
"grad_norm": 0.2925998866558075,
|
| 476 |
+
"learning_rate": 0.00019997478028078853,
|
| 477 |
+
"loss": 0.4643,
|
| 478 |
+
"step": 134
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.07542983915696062,
|
| 482 |
+
"grad_norm": 0.49747079610824585,
|
| 483 |
+
"learning_rate": 0.0001999705839252669,
|
| 484 |
+
"loss": 0.6133,
|
| 485 |
+
"step": 136
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.07653910149750416,
|
| 489 |
+
"grad_norm": 0.37335023283958435,
|
| 490 |
+
"learning_rate": 0.0001999660648388193,
|
| 491 |
+
"loss": 0.508,
|
| 492 |
+
"step": 138
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.0776483638380477,
|
| 496 |
+
"grad_norm": 0.383662611246109,
|
| 497 |
+
"learning_rate": 0.00019996122303603446,
|
| 498 |
+
"loss": 0.5049,
|
| 499 |
+
"step": 140
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.07875762617859124,
|
| 503 |
+
"grad_norm": 0.28099021315574646,
|
| 504 |
+
"learning_rate": 0.000199956058532543,
|
| 505 |
+
"loss": 0.4487,
|
| 506 |
+
"step": 142
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.07986688851913477,
|
| 510 |
+
"grad_norm": 0.2824121415615082,
|
| 511 |
+
"learning_rate": 0.00019995057134501726,
|
| 512 |
+
"loss": 0.4544,
|
| 513 |
+
"step": 144
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.08097615085967831,
|
| 517 |
+
"grad_norm": 0.33720389008522034,
|
| 518 |
+
"learning_rate": 0.00019994476149117133,
|
| 519 |
+
"loss": 0.5987,
|
| 520 |
+
"step": 146
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.08208541320022185,
|
| 524 |
+
"grad_norm": 6.722891330718994,
|
| 525 |
+
"learning_rate": 0.0001999386289897609,
|
| 526 |
+
"loss": 0.705,
|
| 527 |
+
"step": 148
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.08319467554076539,
|
| 531 |
+
"grad_norm": 0.3853927254676819,
|
| 532 |
+
"learning_rate": 0.00019993217386058326,
|
| 533 |
+
"loss": 0.6514,
|
| 534 |
+
"step": 150
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.08430393788130892,
|
| 538 |
+
"grad_norm": 0.32647258043289185,
|
| 539 |
+
"learning_rate": 0.0001999253961244773,
|
| 540 |
+
"loss": 0.4882,
|
| 541 |
+
"step": 152
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.08541320022185246,
|
| 545 |
+
"grad_norm": 0.33191072940826416,
|
| 546 |
+
"learning_rate": 0.0001999182958033232,
|
| 547 |
+
"loss": 0.55,
|
| 548 |
+
"step": 154
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.08652246256239601,
|
| 552 |
+
"grad_norm": 0.3368849456310272,
|
| 553 |
+
"learning_rate": 0.00019991087292004273,
|
| 554 |
+
"loss": 0.4264,
|
| 555 |
+
"step": 156
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.08763172490293955,
|
| 559 |
+
"grad_norm": 0.2839064598083496,
|
| 560 |
+
"learning_rate": 0.00019990312749859886,
|
| 561 |
+
"loss": 0.4845,
|
| 562 |
+
"step": 158
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.08874098724348309,
|
| 566 |
+
"grad_norm": 0.34918418526649475,
|
| 567 |
+
"learning_rate": 0.00019989505956399578,
|
| 568 |
+
"loss": 0.4774,
|
| 569 |
+
"step": 160
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.08985024958402663,
|
| 573 |
+
"grad_norm": 0.5855952501296997,
|
| 574 |
+
"learning_rate": 0.0001998866691422789,
|
| 575 |
+
"loss": 0.5082,
|
| 576 |
+
"step": 162
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.09095951192457016,
|
| 580 |
+
"grad_norm": 0.3832476735115051,
|
| 581 |
+
"learning_rate": 0.00019987795626053468,
|
| 582 |
+
"loss": 0.5372,
|
| 583 |
+
"step": 164
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.0920687742651137,
|
| 587 |
+
"grad_norm": 0.2780646085739136,
|
| 588 |
+
"learning_rate": 0.00019986892094689052,
|
| 589 |
+
"loss": 0.4893,
|
| 590 |
+
"step": 166
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.09317803660565724,
|
| 594 |
+
"grad_norm": 0.3530760109424591,
|
| 595 |
+
"learning_rate": 0.00019985956323051478,
|
| 596 |
+
"loss": 0.4235,
|
| 597 |
+
"step": 168
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.09428729894620078,
|
| 601 |
+
"grad_norm": 0.3618020713329315,
|
| 602 |
+
"learning_rate": 0.00019984988314161658,
|
| 603 |
+
"loss": 0.4818,
|
| 604 |
+
"step": 170
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.09539656128674431,
|
| 608 |
+
"grad_norm": 0.360500693321228,
|
| 609 |
+
"learning_rate": 0.00019983988071144574,
|
| 610 |
+
"loss": 0.602,
|
| 611 |
+
"step": 172
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.09650582362728785,
|
| 615 |
+
"grad_norm": 0.2537604570388794,
|
| 616 |
+
"learning_rate": 0.00019982955597229275,
|
| 617 |
+
"loss": 0.5386,
|
| 618 |
+
"step": 174
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.09761508596783139,
|
| 622 |
+
"grad_norm": 0.31803157925605774,
|
| 623 |
+
"learning_rate": 0.0001998189089574885,
|
| 624 |
+
"loss": 0.4198,
|
| 625 |
+
"step": 176
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.09872434830837493,
|
| 629 |
+
"grad_norm": 0.404077410697937,
|
| 630 |
+
"learning_rate": 0.0001998079397014043,
|
| 631 |
+
"loss": 0.4918,
|
| 632 |
+
"step": 178
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.09983361064891846,
|
| 636 |
+
"grad_norm": 0.3178037703037262,
|
| 637 |
+
"learning_rate": 0.00019979664823945178,
|
| 638 |
+
"loss": 0.5349,
|
| 639 |
+
"step": 180
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.100942872989462,
|
| 643 |
+
"grad_norm": 0.3245227336883545,
|
| 644 |
+
"learning_rate": 0.0001997850346080827,
|
| 645 |
+
"loss": 0.4801,
|
| 646 |
+
"step": 182
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.10205213533000555,
|
| 650 |
+
"grad_norm": 0.33926212787628174,
|
| 651 |
+
"learning_rate": 0.00019977309884478879,
|
| 652 |
+
"loss": 0.4323,
|
| 653 |
+
"step": 184
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.10316139767054909,
|
| 657 |
+
"grad_norm": 0.41488340497016907,
|
| 658 |
+
"learning_rate": 0.0001997608409881019,
|
| 659 |
+
"loss": 0.557,
|
| 660 |
+
"step": 186
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.10427066001109263,
|
| 664 |
+
"grad_norm": 0.40327388048171997,
|
| 665 |
+
"learning_rate": 0.0001997482610775935,
|
| 666 |
+
"loss": 0.4811,
|
| 667 |
+
"step": 188
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.10537992235163617,
|
| 671 |
+
"grad_norm": 0.2919729948043823,
|
| 672 |
+
"learning_rate": 0.0001997353591538748,
|
| 673 |
+
"loss": 0.3776,
|
| 674 |
+
"step": 190
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.1064891846921797,
|
| 678 |
+
"grad_norm": 0.280122846364975,
|
| 679 |
+
"learning_rate": 0.00019972213525859658,
|
| 680 |
+
"loss": 0.4697,
|
| 681 |
+
"step": 192
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.10759844703272324,
|
| 685 |
+
"grad_norm": 0.41433635354042053,
|
| 686 |
+
"learning_rate": 0.00019970858943444897,
|
| 687 |
+
"loss": 0.591,
|
| 688 |
+
"step": 194
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.10870770937326678,
|
| 692 |
+
"grad_norm": 0.33640623092651367,
|
| 693 |
+
"learning_rate": 0.00019969472172516142,
|
| 694 |
+
"loss": 0.523,
|
| 695 |
+
"step": 196
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.10981697171381032,
|
| 699 |
+
"grad_norm": 0.3238784670829773,
|
| 700 |
+
"learning_rate": 0.0001996805321755024,
|
| 701 |
+
"loss": 0.5112,
|
| 702 |
+
"step": 198
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.11092623405435385,
|
| 706 |
+
"grad_norm": 0.33250951766967773,
|
| 707 |
+
"learning_rate": 0.0001996660208312796,
|
| 708 |
+
"loss": 0.4554,
|
| 709 |
+
"step": 200
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.11203549639489739,
|
| 713 |
+
"grad_norm": 0.2348429262638092,
|
| 714 |
+
"learning_rate": 0.0001996511877393393,
|
| 715 |
+
"loss": 0.4454,
|
| 716 |
+
"step": 202
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.11314475873544093,
|
| 720 |
+
"grad_norm": 0.3479189872741699,
|
| 721 |
+
"learning_rate": 0.00019963603294756657,
|
| 722 |
+
"loss": 0.4648,
|
| 723 |
+
"step": 204
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.11425402107598447,
|
| 727 |
+
"grad_norm": 0.2339608371257782,
|
| 728 |
+
"learning_rate": 0.00019962055650488502,
|
| 729 |
+
"loss": 0.4383,
|
| 730 |
+
"step": 206
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.115363283416528,
|
| 734 |
+
"grad_norm": 0.3076520562171936,
|
| 735 |
+
"learning_rate": 0.0001996047584612566,
|
| 736 |
+
"loss": 0.4886,
|
| 737 |
+
"step": 208
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.11647254575707154,
|
| 741 |
+
"grad_norm": 0.3302302658557892,
|
| 742 |
+
"learning_rate": 0.00019958863886768147,
|
| 743 |
+
"loss": 0.5678,
|
| 744 |
+
"step": 210
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.11758180809761509,
|
| 748 |
+
"grad_norm": 0.3337957561016083,
|
| 749 |
+
"learning_rate": 0.00019957219777619786,
|
| 750 |
+
"loss": 0.4573,
|
| 751 |
+
"step": 212
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.11869107043815863,
|
| 755 |
+
"grad_norm": 0.31091246008872986,
|
| 756 |
+
"learning_rate": 0.0001995554352398819,
|
| 757 |
+
"loss": 0.5596,
|
| 758 |
+
"step": 214
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.11980033277870217,
|
| 762 |
+
"grad_norm": 0.27653515338897705,
|
| 763 |
+
"learning_rate": 0.00019953835131284738,
|
| 764 |
+
"loss": 0.4732,
|
| 765 |
+
"step": 216
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.1209095951192457,
|
| 769 |
+
"grad_norm": 0.45777037739753723,
|
| 770 |
+
"learning_rate": 0.00019952094605024562,
|
| 771 |
+
"loss": 0.5798,
|
| 772 |
+
"step": 218
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.12201885745978924,
|
| 776 |
+
"grad_norm": 0.30545416474342346,
|
| 777 |
+
"learning_rate": 0.00019950321950826534,
|
| 778 |
+
"loss": 0.4263,
|
| 779 |
+
"step": 220
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.12312811980033278,
|
| 783 |
+
"grad_norm": 0.29590895771980286,
|
| 784 |
+
"learning_rate": 0.00019948517174413238,
|
| 785 |
+
"loss": 0.4797,
|
| 786 |
+
"step": 222
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.12423738214087632,
|
| 790 |
+
"grad_norm": 0.387412965297699,
|
| 791 |
+
"learning_rate": 0.0001994668028161096,
|
| 792 |
+
"loss": 0.4482,
|
| 793 |
+
"step": 224
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.12534664448141986,
|
| 797 |
+
"grad_norm": 0.337591290473938,
|
| 798 |
+
"learning_rate": 0.00019944811278349667,
|
| 799 |
+
"loss": 0.5011,
|
| 800 |
+
"step": 226
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.1264559068219634,
|
| 804 |
+
"grad_norm": 0.3025722801685333,
|
| 805 |
+
"learning_rate": 0.00019942910170662987,
|
| 806 |
+
"loss": 0.6154,
|
| 807 |
+
"step": 228
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.12756516916250693,
|
| 811 |
+
"grad_norm": 0.2529117465019226,
|
| 812 |
+
"learning_rate": 0.00019940976964688182,
|
| 813 |
+
"loss": 0.4412,
|
| 814 |
+
"step": 230
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.12867443150305047,
|
| 818 |
+
"grad_norm": 0.3442631959915161,
|
| 819 |
+
"learning_rate": 0.0001993901166666615,
|
| 820 |
+
"loss": 0.5209,
|
| 821 |
+
"step": 232
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.129783693843594,
|
| 825 |
+
"grad_norm": 0.30931201577186584,
|
| 826 |
+
"learning_rate": 0.00019937014282941373,
|
| 827 |
+
"loss": 0.6154,
|
| 828 |
+
"step": 234
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.13089295618413754,
|
| 832 |
+
"grad_norm": 0.3555050790309906,
|
| 833 |
+
"learning_rate": 0.00019934984819961927,
|
| 834 |
+
"loss": 0.5173,
|
| 835 |
+
"step": 236
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.13200221852468108,
|
| 839 |
+
"grad_norm": 0.44570550322532654,
|
| 840 |
+
"learning_rate": 0.00019932923284279446,
|
| 841 |
+
"loss": 0.5319,
|
| 842 |
+
"step": 238
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.13311148086522462,
|
| 846 |
+
"grad_norm": 0.3192990720272064,
|
| 847 |
+
"learning_rate": 0.00019930829682549095,
|
| 848 |
+
"loss": 0.581,
|
| 849 |
+
"step": 240
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.13422074320576816,
|
| 853 |
+
"grad_norm": 0.24486322700977325,
|
| 854 |
+
"learning_rate": 0.00019928704021529567,
|
| 855 |
+
"loss": 0.403,
|
| 856 |
+
"step": 242
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.1353300055463117,
|
| 860 |
+
"grad_norm": 0.2888510823249817,
|
| 861 |
+
"learning_rate": 0.00019926546308083047,
|
| 862 |
+
"loss": 0.4869,
|
| 863 |
+
"step": 244
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.13643926788685523,
|
| 867 |
+
"grad_norm": 0.3472958207130432,
|
| 868 |
+
"learning_rate": 0.00019924356549175188,
|
| 869 |
+
"loss": 0.6019,
|
| 870 |
+
"step": 246
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.13754853022739877,
|
| 874 |
+
"grad_norm": 0.2682150900363922,
|
| 875 |
+
"learning_rate": 0.00019922134751875102,
|
| 876 |
+
"loss": 0.4816,
|
| 877 |
+
"step": 248
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.1386577925679423,
|
| 881 |
+
"grad_norm": 0.24147269129753113,
|
| 882 |
+
"learning_rate": 0.00019919880923355323,
|
| 883 |
+
"loss": 0.4681,
|
| 884 |
+
"step": 250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.13976705490848584,
|
| 888 |
+
"grad_norm": 0.3548056185245514,
|
| 889 |
+
"learning_rate": 0.00019917595070891798,
|
| 890 |
+
"loss": 0.568,
|
| 891 |
+
"step": 252
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.1408763172490294,
|
| 895 |
+
"grad_norm": 0.299034982919693,
|
| 896 |
+
"learning_rate": 0.00019915277201863844,
|
| 897 |
+
"loss": 0.4479,
|
| 898 |
+
"step": 254
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.14198557958957295,
|
| 902 |
+
"grad_norm": 0.32234734296798706,
|
| 903 |
+
"learning_rate": 0.00019912927323754146,
|
| 904 |
+
"loss": 0.542,
|
| 905 |
+
"step": 256
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.14309484193011648,
|
| 909 |
+
"grad_norm": 0.31738007068634033,
|
| 910 |
+
"learning_rate": 0.00019910545444148722,
|
| 911 |
+
"loss": 0.4458,
|
| 912 |
+
"step": 258
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.14420410427066002,
|
| 916 |
+
"grad_norm": 0.2500552833080292,
|
| 917 |
+
"learning_rate": 0.0001990813157073689,
|
| 918 |
+
"loss": 0.4774,
|
| 919 |
+
"step": 260
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.14531336661120356,
|
| 923 |
+
"grad_norm": 0.2912147641181946,
|
| 924 |
+
"learning_rate": 0.0001990568571131126,
|
| 925 |
+
"loss": 0.523,
|
| 926 |
+
"step": 262
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.1464226289517471,
|
| 930 |
+
"grad_norm": 0.35038962960243225,
|
| 931 |
+
"learning_rate": 0.00019903207873767705,
|
| 932 |
+
"loss": 0.5325,
|
| 933 |
+
"step": 264
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.14753189129229063,
|
| 937 |
+
"grad_norm": 0.6067700386047363,
|
| 938 |
+
"learning_rate": 0.00019900698066105317,
|
| 939 |
+
"loss": 0.5304,
|
| 940 |
+
"step": 266
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.14864115363283417,
|
| 944 |
+
"grad_norm": 0.8355000615119934,
|
| 945 |
+
"learning_rate": 0.00019898156296426414,
|
| 946 |
+
"loss": 0.6244,
|
| 947 |
+
"step": 268
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.1497504159733777,
|
| 951 |
+
"grad_norm": 0.2898505926132202,
|
| 952 |
+
"learning_rate": 0.00019895582572936475,
|
| 953 |
+
"loss": 0.4361,
|
| 954 |
+
"step": 270
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.15085967831392125,
|
| 958 |
+
"grad_norm": 0.2776695191860199,
|
| 959 |
+
"learning_rate": 0.0001989297690394416,
|
| 960 |
+
"loss": 0.5019,
|
| 961 |
+
"step": 272
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.15196894065446478,
|
| 965 |
+
"grad_norm": 0.310939222574234,
|
| 966 |
+
"learning_rate": 0.0001989033929786123,
|
| 967 |
+
"loss": 0.44,
|
| 968 |
+
"step": 274
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.15307820299500832,
|
| 972 |
+
"grad_norm": 0.37729912996292114,
|
| 973 |
+
"learning_rate": 0.00019887669763202567,
|
| 974 |
+
"loss": 0.6571,
|
| 975 |
+
"step": 276
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.15418746533555186,
|
| 979 |
+
"grad_norm": 0.3895605802536011,
|
| 980 |
+
"learning_rate": 0.0001988496830858612,
|
| 981 |
+
"loss": 0.4787,
|
| 982 |
+
"step": 278
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.1552967276760954,
|
| 986 |
+
"grad_norm": 0.27666375041007996,
|
| 987 |
+
"learning_rate": 0.00019882234942732882,
|
| 988 |
+
"loss": 0.4677,
|
| 989 |
+
"step": 280
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.15640599001663893,
|
| 993 |
+
"grad_norm": 0.5543416142463684,
|
| 994 |
+
"learning_rate": 0.00019879469674466868,
|
| 995 |
+
"loss": 0.5272,
|
| 996 |
+
"step": 282
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.15751525235718247,
|
| 1000 |
+
"grad_norm": 0.46799716353416443,
|
| 1001 |
+
"learning_rate": 0.00019876672512715078,
|
| 1002 |
+
"loss": 0.3936,
|
| 1003 |
+
"step": 284
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.158624514697726,
|
| 1007 |
+
"grad_norm": 0.47091013193130493,
|
| 1008 |
+
"learning_rate": 0.00019873843466507475,
|
| 1009 |
+
"loss": 0.5349,
|
| 1010 |
+
"step": 286
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.15973377703826955,
|
| 1014 |
+
"grad_norm": 0.2690078020095825,
|
| 1015 |
+
"learning_rate": 0.0001987098254497695,
|
| 1016 |
+
"loss": 0.4377,
|
| 1017 |
+
"step": 288
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.16084303937881309,
|
| 1021 |
+
"grad_norm": 0.8880047798156738,
|
| 1022 |
+
"learning_rate": 0.000198680897573593,
|
| 1023 |
+
"loss": 0.4537,
|
| 1024 |
+
"step": 290
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.16195230171935662,
|
| 1028 |
+
"grad_norm": 0.25852563977241516,
|
| 1029 |
+
"learning_rate": 0.00019865165112993195,
|
| 1030 |
+
"loss": 0.4001,
|
| 1031 |
+
"step": 292
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.16306156405990016,
|
| 1035 |
+
"grad_norm": 0.3986980617046356,
|
| 1036 |
+
"learning_rate": 0.00019862208621320142,
|
| 1037 |
+
"loss": 0.5538,
|
| 1038 |
+
"step": 294
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.1641708264004437,
|
| 1042 |
+
"grad_norm": 0.2550656795501709,
|
| 1043 |
+
"learning_rate": 0.00019859220291884458,
|
| 1044 |
+
"loss": 0.4261,
|
| 1045 |
+
"step": 296
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.16528008874098724,
|
| 1049 |
+
"grad_norm": 0.2604806125164032,
|
| 1050 |
+
"learning_rate": 0.0001985620013433325,
|
| 1051 |
+
"loss": 0.5003,
|
| 1052 |
+
"step": 298
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.16638935108153077,
|
| 1056 |
+
"grad_norm": 0.3848886787891388,
|
| 1057 |
+
"learning_rate": 0.0001985314815841637,
|
| 1058 |
+
"loss": 0.4393,
|
| 1059 |
+
"step": 300
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.1674986134220743,
|
| 1063 |
+
"grad_norm": 0.37041744589805603,
|
| 1064 |
+
"learning_rate": 0.00019850064373986377,
|
| 1065 |
+
"loss": 0.4831,
|
| 1066 |
+
"step": 302
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.16860787576261785,
|
| 1070 |
+
"grad_norm": 0.4813826382160187,
|
| 1071 |
+
"learning_rate": 0.0001984694879099853,
|
| 1072 |
+
"loss": 0.5145,
|
| 1073 |
+
"step": 304
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.16971713810316139,
|
| 1077 |
+
"grad_norm": 0.3015052080154419,
|
| 1078 |
+
"learning_rate": 0.00019843801419510744,
|
| 1079 |
+
"loss": 0.4991,
|
| 1080 |
+
"step": 306
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.17082640044370492,
|
| 1084 |
+
"grad_norm": 0.33228328824043274,
|
| 1085 |
+
"learning_rate": 0.00019840622269683538,
|
| 1086 |
+
"loss": 0.4475,
|
| 1087 |
+
"step": 308
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.1719356627842485,
|
| 1091 |
+
"grad_norm": 0.3386059105396271,
|
| 1092 |
+
"learning_rate": 0.00019837411351780038,
|
| 1093 |
+
"loss": 0.5565,
|
| 1094 |
+
"step": 310
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.17304492512479203,
|
| 1098 |
+
"grad_norm": 0.38623613119125366,
|
| 1099 |
+
"learning_rate": 0.00019834168676165917,
|
| 1100 |
+
"loss": 0.5547,
|
| 1101 |
+
"step": 312
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.17415418746533556,
|
| 1105 |
+
"grad_norm": 0.32846319675445557,
|
| 1106 |
+
"learning_rate": 0.0001983089425330937,
|
| 1107 |
+
"loss": 0.547,
|
| 1108 |
+
"step": 314
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.1752634498058791,
|
| 1112 |
+
"grad_norm": 0.3607962429523468,
|
| 1113 |
+
"learning_rate": 0.00019827588093781083,
|
| 1114 |
+
"loss": 0.5914,
|
| 1115 |
+
"step": 316
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.17637271214642264,
|
| 1119 |
+
"grad_norm": 0.36206066608428955,
|
| 1120 |
+
"learning_rate": 0.00019824250208254194,
|
| 1121 |
+
"loss": 0.4694,
|
| 1122 |
+
"step": 318
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.17748197448696618,
|
| 1126 |
+
"grad_norm": 0.24287384748458862,
|
| 1127 |
+
"learning_rate": 0.0001982088060750426,
|
| 1128 |
+
"loss": 0.3787,
|
| 1129 |
+
"step": 320
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.17859123682750971,
|
| 1133 |
+
"grad_norm": 0.35104668140411377,
|
| 1134 |
+
"learning_rate": 0.00019817479302409227,
|
| 1135 |
+
"loss": 0.4743,
|
| 1136 |
+
"step": 322
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.17970049916805325,
|
| 1140 |
+
"grad_norm": 0.29966557025909424,
|
| 1141 |
+
"learning_rate": 0.0001981404630394939,
|
| 1142 |
+
"loss": 0.5124,
|
| 1143 |
+
"step": 324
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.1808097615085968,
|
| 1147 |
+
"grad_norm": 0.35917168855667114,
|
| 1148 |
+
"learning_rate": 0.0001981058162320735,
|
| 1149 |
+
"loss": 0.5524,
|
| 1150 |
+
"step": 326
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.18191902384914033,
|
| 1154 |
+
"grad_norm": 0.38302454352378845,
|
| 1155 |
+
"learning_rate": 0.00019807085271368005,
|
| 1156 |
+
"loss": 0.6019,
|
| 1157 |
+
"step": 328
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.18302828618968386,
|
| 1161 |
+
"grad_norm": 0.24732793867588043,
|
| 1162 |
+
"learning_rate": 0.0001980355725971847,
|
| 1163 |
+
"loss": 0.4952,
|
| 1164 |
+
"step": 330
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.1841375485302274,
|
| 1168 |
+
"grad_norm": 0.279240220785141,
|
| 1169 |
+
"learning_rate": 0.0001979999759964809,
|
| 1170 |
+
"loss": 0.3949,
|
| 1171 |
+
"step": 332
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.18524681087077094,
|
| 1175 |
+
"grad_norm": 0.2902393341064453,
|
| 1176 |
+
"learning_rate": 0.00019796406302648368,
|
| 1177 |
+
"loss": 0.4938,
|
| 1178 |
+
"step": 334
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.18635607321131448,
|
| 1182 |
+
"grad_norm": 0.3428916037082672,
|
| 1183 |
+
"learning_rate": 0.00019792783380312936,
|
| 1184 |
+
"loss": 0.4401,
|
| 1185 |
+
"step": 336
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.18746533555185801,
|
| 1189 |
+
"grad_norm": 0.38825634121894836,
|
| 1190 |
+
"learning_rate": 0.00019789128844337528,
|
| 1191 |
+
"loss": 0.5968,
|
| 1192 |
+
"step": 338
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.18857459789240155,
|
| 1196 |
+
"grad_norm": 0.26601892709732056,
|
| 1197 |
+
"learning_rate": 0.0001978544270651993,
|
| 1198 |
+
"loss": 0.3993,
|
| 1199 |
+
"step": 340
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.1896838602329451,
|
| 1203 |
+
"grad_norm": 0.29738616943359375,
|
| 1204 |
+
"learning_rate": 0.00019781724978759955,
|
| 1205 |
+
"loss": 0.4975,
|
| 1206 |
+
"step": 342
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.19079312257348863,
|
| 1210 |
+
"grad_norm": 0.3405514061450958,
|
| 1211 |
+
"learning_rate": 0.00019777975673059383,
|
| 1212 |
+
"loss": 0.5674,
|
| 1213 |
+
"step": 344
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.19190238491403216,
|
| 1217 |
+
"grad_norm": 0.29235440492630005,
|
| 1218 |
+
"learning_rate": 0.00019774194801521947,
|
| 1219 |
+
"loss": 0.4854,
|
| 1220 |
+
"step": 346
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.1930116472545757,
|
| 1224 |
+
"grad_norm": 0.279852032661438,
|
| 1225 |
+
"learning_rate": 0.00019770382376353284,
|
| 1226 |
+
"loss": 0.3933,
|
| 1227 |
+
"step": 348
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.19412090959511924,
|
| 1231 |
+
"grad_norm": 0.3054843544960022,
|
| 1232 |
+
"learning_rate": 0.00019766538409860882,
|
| 1233 |
+
"loss": 0.5125,
|
| 1234 |
+
"step": 350
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.19523017193566278,
|
| 1238 |
+
"grad_norm": 0.2959388494491577,
|
| 1239 |
+
"learning_rate": 0.00019762662914454065,
|
| 1240 |
+
"loss": 0.5114,
|
| 1241 |
+
"step": 352
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.19633943427620631,
|
| 1245 |
+
"grad_norm": 0.42934173345565796,
|
| 1246 |
+
"learning_rate": 0.0001975875590264393,
|
| 1247 |
+
"loss": 0.5157,
|
| 1248 |
+
"step": 354
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.19744869661674985,
|
| 1252 |
+
"grad_norm": 0.28414642810821533,
|
| 1253 |
+
"learning_rate": 0.00019754817387043327,
|
| 1254 |
+
"loss": 0.4952,
|
| 1255 |
+
"step": 356
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.1985579589572934,
|
| 1259 |
+
"grad_norm": 0.27035775780677795,
|
| 1260 |
+
"learning_rate": 0.00019750847380366806,
|
| 1261 |
+
"loss": 0.3946,
|
| 1262 |
+
"step": 358
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.19966722129783693,
|
| 1266 |
+
"grad_norm": 0.23628903925418854,
|
| 1267 |
+
"learning_rate": 0.0001974684589543057,
|
| 1268 |
+
"loss": 0.3691,
|
| 1269 |
+
"step": 360
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.20077648363838047,
|
| 1273 |
+
"grad_norm": 0.3877003788948059,
|
| 1274 |
+
"learning_rate": 0.0001974281294515245,
|
| 1275 |
+
"loss": 0.5729,
|
| 1276 |
+
"step": 362
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.201885745978924,
|
| 1280 |
+
"grad_norm": 0.3051539659500122,
|
| 1281 |
+
"learning_rate": 0.00019738748542551861,
|
| 1282 |
+
"loss": 0.4378,
|
| 1283 |
+
"step": 364
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.20299500831946754,
|
| 1287 |
+
"grad_norm": 0.27406755089759827,
|
| 1288 |
+
"learning_rate": 0.00019734652700749737,
|
| 1289 |
+
"loss": 0.505,
|
| 1290 |
+
"step": 366
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.2041042706600111,
|
| 1294 |
+
"grad_norm": 2.0898008346557617,
|
| 1295 |
+
"learning_rate": 0.0001973052543296852,
|
| 1296 |
+
"loss": 0.5647,
|
| 1297 |
+
"step": 368
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.20521353300055464,
|
| 1301 |
+
"grad_norm": 0.24537453055381775,
|
| 1302 |
+
"learning_rate": 0.000197263667525321,
|
| 1303 |
+
"loss": 0.4543,
|
| 1304 |
+
"step": 370
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.20632279534109818,
|
| 1308 |
+
"grad_norm": 0.26961207389831543,
|
| 1309 |
+
"learning_rate": 0.0001972217667286577,
|
| 1310 |
+
"loss": 0.4595,
|
| 1311 |
+
"step": 372
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.20743205768164172,
|
| 1315 |
+
"grad_norm": 0.24330930411815643,
|
| 1316 |
+
"learning_rate": 0.00019717955207496196,
|
| 1317 |
+
"loss": 0.4383,
|
| 1318 |
+
"step": 374
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.20854132002218526,
|
| 1322 |
+
"grad_norm": 0.42495131492614746,
|
| 1323 |
+
"learning_rate": 0.0001971370237005136,
|
| 1324 |
+
"loss": 0.6703,
|
| 1325 |
+
"step": 376
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.2096505823627288,
|
| 1329 |
+
"grad_norm": 0.28645631670951843,
|
| 1330 |
+
"learning_rate": 0.0001970941817426052,
|
| 1331 |
+
"loss": 0.4693,
|
| 1332 |
+
"step": 378
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.21075984470327233,
|
| 1336 |
+
"grad_norm": 0.27761217951774597,
|
| 1337 |
+
"learning_rate": 0.00019705102633954172,
|
| 1338 |
+
"loss": 0.5217,
|
| 1339 |
+
"step": 380
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.21186910704381587,
|
| 1343 |
+
"grad_norm": 0.27868810296058655,
|
| 1344 |
+
"learning_rate": 0.00019700755763063998,
|
| 1345 |
+
"loss": 0.4173,
|
| 1346 |
+
"step": 382
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.2129783693843594,
|
| 1350 |
+
"grad_norm": 0.3163412809371948,
|
| 1351 |
+
"learning_rate": 0.0001969637757562282,
|
| 1352 |
+
"loss": 0.5144,
|
| 1353 |
+
"step": 384
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.21408763172490294,
|
| 1357 |
+
"grad_norm": 0.2901851534843445,
|
| 1358 |
+
"learning_rate": 0.00019691968085764562,
|
| 1359 |
+
"loss": 0.5128,
|
| 1360 |
+
"step": 386
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.21519689406544648,
|
| 1364 |
+
"grad_norm": 0.3356168270111084,
|
| 1365 |
+
"learning_rate": 0.00019687527307724197,
|
| 1366 |
+
"loss": 0.4398,
|
| 1367 |
+
"step": 388
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.21630615640599002,
|
| 1371 |
+
"grad_norm": 0.33736830949783325,
|
| 1372 |
+
"learning_rate": 0.0001968305525583771,
|
| 1373 |
+
"loss": 0.4771,
|
| 1374 |
+
"step": 390
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.21741541874653356,
|
| 1378 |
+
"grad_norm": 0.2211776226758957,
|
| 1379 |
+
"learning_rate": 0.00019678551944542037,
|
| 1380 |
+
"loss": 0.3975,
|
| 1381 |
+
"step": 392
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.2185246810870771,
|
| 1385 |
+
"grad_norm": 0.2705392837524414,
|
| 1386 |
+
"learning_rate": 0.00019674017388375038,
|
| 1387 |
+
"loss": 0.4347,
|
| 1388 |
+
"step": 394
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.21963394342762063,
|
| 1392 |
+
"grad_norm": 0.2636098861694336,
|
| 1393 |
+
"learning_rate": 0.0001966945160197543,
|
| 1394 |
+
"loss": 0.5036,
|
| 1395 |
+
"step": 396
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.22074320576816417,
|
| 1399 |
+
"grad_norm": 0.26003068685531616,
|
| 1400 |
+
"learning_rate": 0.00019664854600082756,
|
| 1401 |
+
"loss": 0.458,
|
| 1402 |
+
"step": 398
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.2218524681087077,
|
| 1406 |
+
"grad_norm": 0.1831740140914917,
|
| 1407 |
+
"learning_rate": 0.00019660226397537326,
|
| 1408 |
+
"loss": 0.4322,
|
| 1409 |
+
"step": 400
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.22296173044925124,
|
| 1413 |
+
"grad_norm": 1.4381052255630493,
|
| 1414 |
+
"learning_rate": 0.00019655567009280178,
|
| 1415 |
+
"loss": 0.545,
|
| 1416 |
+
"step": 402
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.22407099278979478,
|
| 1420 |
+
"grad_norm": 0.3401663303375244,
|
| 1421 |
+
"learning_rate": 0.00019650876450353022,
|
| 1422 |
+
"loss": 0.5524,
|
| 1423 |
+
"step": 404
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.22518025513033832,
|
| 1427 |
+
"grad_norm": 0.294527530670166,
|
| 1428 |
+
"learning_rate": 0.00019646154735898202,
|
| 1429 |
+
"loss": 0.4059,
|
| 1430 |
+
"step": 406
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.22628951747088186,
|
| 1434 |
+
"grad_norm": 0.3033091127872467,
|
| 1435 |
+
"learning_rate": 0.00019641401881158625,
|
| 1436 |
+
"loss": 0.4119,
|
| 1437 |
+
"step": 408
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.2273987798114254,
|
| 1441 |
+
"grad_norm": 0.3446100950241089,
|
| 1442 |
+
"learning_rate": 0.00019636617901477746,
|
| 1443 |
+
"loss": 0.4299,
|
| 1444 |
+
"step": 410
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.22850804215196893,
|
| 1448 |
+
"grad_norm": 0.2955077886581421,
|
| 1449 |
+
"learning_rate": 0.00019631802812299483,
|
| 1450 |
+
"loss": 0.4496,
|
| 1451 |
+
"step": 412
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.22961730449251247,
|
| 1455 |
+
"grad_norm": 0.2663453221321106,
|
| 1456 |
+
"learning_rate": 0.00019626956629168192,
|
| 1457 |
+
"loss": 0.5373,
|
| 1458 |
+
"step": 414
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.230726566833056,
|
| 1462 |
+
"grad_norm": 0.31393304467201233,
|
| 1463 |
+
"learning_rate": 0.0001962207936772861,
|
| 1464 |
+
"loss": 0.4856,
|
| 1465 |
+
"step": 416
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.23183582917359954,
|
| 1469 |
+
"grad_norm": 0.3716754615306854,
|
| 1470 |
+
"learning_rate": 0.00019617171043725796,
|
| 1471 |
+
"loss": 0.6161,
|
| 1472 |
+
"step": 418
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.23294509151414308,
|
| 1476 |
+
"grad_norm": 0.3159737288951874,
|
| 1477 |
+
"learning_rate": 0.00019612231673005092,
|
| 1478 |
+
"loss": 0.4493,
|
| 1479 |
+
"step": 420
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.23405435385468662,
|
| 1483 |
+
"grad_norm": 0.3315045237541199,
|
| 1484 |
+
"learning_rate": 0.00019607261271512068,
|
| 1485 |
+
"loss": 0.4695,
|
| 1486 |
+
"step": 422
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.23516361619523019,
|
| 1490 |
+
"grad_norm": 0.27597951889038086,
|
| 1491 |
+
"learning_rate": 0.0001960225985529246,
|
| 1492 |
+
"loss": 0.3391,
|
| 1493 |
+
"step": 424
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.23627287853577372,
|
| 1497 |
+
"grad_norm": 0.4410407543182373,
|
| 1498 |
+
"learning_rate": 0.00019597227440492143,
|
| 1499 |
+
"loss": 0.5034,
|
| 1500 |
+
"step": 426
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.23738214087631726,
|
| 1504 |
+
"grad_norm": 0.2749500572681427,
|
| 1505 |
+
"learning_rate": 0.00019592164043357046,
|
| 1506 |
+
"loss": 0.4934,
|
| 1507 |
+
"step": 428
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.2384914032168608,
|
| 1511 |
+
"grad_norm": 0.32692408561706543,
|
| 1512 |
+
"learning_rate": 0.00019587069680233134,
|
| 1513 |
+
"loss": 0.4589,
|
| 1514 |
+
"step": 430
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.23960066555740434,
|
| 1518 |
+
"grad_norm": 0.2124568372964859,
|
| 1519 |
+
"learning_rate": 0.00019581944367566326,
|
| 1520 |
+
"loss": 0.4367,
|
| 1521 |
+
"step": 432
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.24070992789794787,
|
| 1525 |
+
"grad_norm": 0.26193922758102417,
|
| 1526 |
+
"learning_rate": 0.00019576788121902457,
|
| 1527 |
+
"loss": 0.4172,
|
| 1528 |
+
"step": 434
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.2418191902384914,
|
| 1532 |
+
"grad_norm": 0.25930801033973694,
|
| 1533 |
+
"learning_rate": 0.00019571600959887223,
|
| 1534 |
+
"loss": 0.3983,
|
| 1535 |
+
"step": 436
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.24292845257903495,
|
| 1539 |
+
"grad_norm": 0.30770912766456604,
|
| 1540 |
+
"learning_rate": 0.0001956638289826613,
|
| 1541 |
+
"loss": 0.4245,
|
| 1542 |
+
"step": 438
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.24403771491957849,
|
| 1546 |
+
"grad_norm": 0.26078179478645325,
|
| 1547 |
+
"learning_rate": 0.00019561133953884427,
|
| 1548 |
+
"loss": 0.3807,
|
| 1549 |
+
"step": 440
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.24514697726012202,
|
| 1553 |
+
"grad_norm": 0.3434695303440094,
|
| 1554 |
+
"learning_rate": 0.00019555854143687068,
|
| 1555 |
+
"loss": 0.6136,
|
| 1556 |
+
"step": 442
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.24625623960066556,
|
| 1560 |
+
"grad_norm": 0.3655042350292206,
|
| 1561 |
+
"learning_rate": 0.00019550543484718648,
|
| 1562 |
+
"loss": 0.4969,
|
| 1563 |
+
"step": 444
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.2473655019412091,
|
| 1567 |
+
"grad_norm": 0.2804737389087677,
|
| 1568 |
+
"learning_rate": 0.00019545201994123344,
|
| 1569 |
+
"loss": 0.4154,
|
| 1570 |
+
"step": 446
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.24847476428175264,
|
| 1574 |
+
"grad_norm": 0.23757784068584442,
|
| 1575 |
+
"learning_rate": 0.00019539829689144876,
|
| 1576 |
+
"loss": 0.467,
|
| 1577 |
+
"step": 448
|
| 1578 |
}
|
| 1579 |
],
|
| 1580 |
"logging_steps": 2,
|
| 1581 |
"max_steps": 3606,
|
| 1582 |
"num_input_tokens_seen": 0,
|
| 1583 |
"num_train_epochs": 2,
|
| 1584 |
+
"save_steps": 64,
|
| 1585 |
"stateful_callbacks": {
|
| 1586 |
"TrainerControl": {
|
| 1587 |
"args": {
|
|
|
|
| 1594 |
"attributes": {}
|
| 1595 |
}
|
| 1596 |
},
|
| 1597 |
+
"total_flos": 1091252539883520.0,
|
| 1598 |
"train_batch_size": 8,
|
| 1599 |
"trial_name": null,
|
| 1600 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81deb56b5e82d7001bcc888a41a74fc8c2ddc0612f58053768650f477b3da027
|
| 3 |
size 5304
|