Training in progress, step 452, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2503003904
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1398a514e57c9606f1325d1ece164366fdd869fd92ee72b36ebabf0478395ec4
|
| 3 |
size 2503003904
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5006244836
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92f271f102ad2c48950c7271225490bd5da827a49c1d29c8a1f17663202f8611
|
| 3 |
size 5006244836
|
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:a2929084b7fe63716f37993380118d101fad7e11dc07866aca47877be7d0e8b7
|
| 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:da6e8c29f5e1878c902dc064fffbc0de189a6a4d4474a3b333a3d28aae6c8698
|
| 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,
|
|
@@ -1680,6 +1680,1497 @@
|
|
| 1680 |
"learning_rate": 6.621275746164437e-05,
|
| 1681 |
"loss": 4.599,
|
| 1682 |
"step": 239
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1683 |
}
|
| 1684 |
],
|
| 1685 |
"logging_steps": 1,
|
|
@@ -1694,12 +3185,12 @@
|
|
| 1694 |
"should_evaluate": false,
|
| 1695 |
"should_log": false,
|
| 1696 |
"should_save": true,
|
| 1697 |
-
"should_training_stop":
|
| 1698 |
},
|
| 1699 |
"attributes": {}
|
| 1700 |
}
|
| 1701 |
},
|
| 1702 |
-
"total_flos":
|
| 1703 |
"train_batch_size": 4,
|
| 1704 |
"trial_name": null,
|
| 1705 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9983434566537824,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 452,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1680 |
"learning_rate": 6.621275746164437e-05,
|
| 1681 |
"loss": 4.599,
|
| 1682 |
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.530093870789619,
|
| 1686 |
+
"grad_norm": 9.09389591217041,
|
| 1687 |
+
"learning_rate": 6.578997938075125e-05,
|
| 1688 |
+
"loss": 4.443,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.5323025952512425,
|
| 1693 |
+
"grad_norm": 8.505534172058105,
|
| 1694 |
+
"learning_rate": 6.536594355341305e-05,
|
| 1695 |
+
"loss": 4.5598,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.5345113197128658,
|
| 1700 |
+
"grad_norm": 9.29316520690918,
|
| 1701 |
+
"learning_rate": 6.494068375608646e-05,
|
| 1702 |
+
"loss": 4.5639,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.5367200441744893,
|
| 1707 |
+
"grad_norm": 8.853930473327637,
|
| 1708 |
+
"learning_rate": 6.451423386272312e-05,
|
| 1709 |
+
"loss": 4.6652,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.5389287686361126,
|
| 1714 |
+
"grad_norm": 9.70859146118164,
|
| 1715 |
+
"learning_rate": 6.408662784207149e-05,
|
| 1716 |
+
"loss": 5.0798,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.5411374930977361,
|
| 1721 |
+
"grad_norm": 9.851663589477539,
|
| 1722 |
+
"learning_rate": 6.3657899754971e-05,
|
| 1723 |
+
"loss": 4.6318,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.5433462175593595,
|
| 1728 |
+
"grad_norm": 9.574539184570312,
|
| 1729 |
+
"learning_rate": 6.322808375163895e-05,
|
| 1730 |
+
"loss": 4.541,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.5455549420209829,
|
| 1735 |
+
"grad_norm": 10.078091621398926,
|
| 1736 |
+
"learning_rate": 6.279721406895038e-05,
|
| 1737 |
+
"loss": 4.6311,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.5477636664826063,
|
| 1742 |
+
"grad_norm": 11.278042793273926,
|
| 1743 |
+
"learning_rate": 6.236532502771078e-05,
|
| 1744 |
+
"loss": 5.0703,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5499723909442297,
|
| 1749 |
+
"grad_norm": 10.999835968017578,
|
| 1750 |
+
"learning_rate": 6.193245102992244e-05,
|
| 1751 |
+
"loss": 5.322,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.5521811154058531,
|
| 1756 |
+
"grad_norm": 10.701493263244629,
|
| 1757 |
+
"learning_rate": 6.149862655604404e-05,
|
| 1758 |
+
"loss": 4.7289,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5543898398674766,
|
| 1763 |
+
"grad_norm": 4.466903209686279,
|
| 1764 |
+
"learning_rate": 6.106388616224419e-05,
|
| 1765 |
+
"loss": 3.7511,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.5565985643290999,
|
| 1770 |
+
"grad_norm": 6.9619526863098145,
|
| 1771 |
+
"learning_rate": 6.062826447764883e-05,
|
| 1772 |
+
"loss": 4.6559,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5588072887907234,
|
| 1777 |
+
"grad_norm": 6.754791736602783,
|
| 1778 |
+
"learning_rate": 6.019179620158294e-05,
|
| 1779 |
+
"loss": 4.7606,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5610160132523467,
|
| 1784 |
+
"grad_norm": 7.005521774291992,
|
| 1785 |
+
"learning_rate": 5.9754516100806423e-05,
|
| 1786 |
+
"loss": 4.4367,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.5632247377139702,
|
| 1791 |
+
"grad_norm": 7.08423376083374,
|
| 1792 |
+
"learning_rate": 5.93164590067449e-05,
|
| 1793 |
+
"loss": 4.5616,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.5654334621755936,
|
| 1798 |
+
"grad_norm": 6.051517009735107,
|
| 1799 |
+
"learning_rate": 5.887765981271518e-05,
|
| 1800 |
+
"loss": 4.6624,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.567642186637217,
|
| 1805 |
+
"grad_norm": 6.105447292327881,
|
| 1806 |
+
"learning_rate": 5.84381534711458e-05,
|
| 1807 |
+
"loss": 4.5011,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.5698509110988405,
|
| 1812 |
+
"grad_norm": 6.139936447143555,
|
| 1813 |
+
"learning_rate": 5.799797499079301e-05,
|
| 1814 |
+
"loss": 4.5359,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.5720596355604638,
|
| 1819 |
+
"grad_norm": 5.590278148651123,
|
| 1820 |
+
"learning_rate": 5.7557159433952034e-05,
|
| 1821 |
+
"loss": 4.6673,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.5742683600220873,
|
| 1826 |
+
"grad_norm": 5.5569586753845215,
|
| 1827 |
+
"learning_rate": 5.7115741913664264e-05,
|
| 1828 |
+
"loss": 4.2704,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.5764770844837107,
|
| 1833 |
+
"grad_norm": 5.2458319664001465,
|
| 1834 |
+
"learning_rate": 5.6673757590920306e-05,
|
| 1835 |
+
"loss": 4.0604,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.5786858089453341,
|
| 1840 |
+
"grad_norm": 5.579367637634277,
|
| 1841 |
+
"learning_rate": 5.62312416718593e-05,
|
| 1842 |
+
"loss": 4.4609,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.5808945334069575,
|
| 1847 |
+
"grad_norm": 5.590094566345215,
|
| 1848 |
+
"learning_rate": 5.578822940496451e-05,
|
| 1849 |
+
"loss": 4.7488,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.5831032578685809,
|
| 1854 |
+
"grad_norm": 5.870893955230713,
|
| 1855 |
+
"learning_rate": 5.534475607825566e-05,
|
| 1856 |
+
"loss": 4.5398,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.5853119823302043,
|
| 1861 |
+
"grad_norm": 6.134853839874268,
|
| 1862 |
+
"learning_rate": 5.490085701647805e-05,
|
| 1863 |
+
"loss": 4.6951,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.5875207067918278,
|
| 1868 |
+
"grad_norm": 6.254935264587402,
|
| 1869 |
+
"learning_rate": 5.44565675782888e-05,
|
| 1870 |
+
"loss": 4.5561,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.5897294312534511,
|
| 1875 |
+
"grad_norm": 6.584839344024658,
|
| 1876 |
+
"learning_rate": 5.401192315344039e-05,
|
| 1877 |
+
"loss": 4.8534,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.5919381557150746,
|
| 1882 |
+
"grad_norm": 7.013578414916992,
|
| 1883 |
+
"learning_rate": 5.3566959159961615e-05,
|
| 1884 |
+
"loss": 4.6348,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.5941468801766979,
|
| 1889 |
+
"grad_norm": 7.184816837310791,
|
| 1890 |
+
"learning_rate": 5.312171104133646e-05,
|
| 1891 |
+
"loss": 4.271,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.5963556046383214,
|
| 1896 |
+
"grad_norm": 7.353315830230713,
|
| 1897 |
+
"learning_rate": 5.2676214263680754e-05,
|
| 1898 |
+
"loss": 4.462,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.5985643290999448,
|
| 1903 |
+
"grad_norm": 8.102008819580078,
|
| 1904 |
+
"learning_rate": 5.223050431291729e-05,
|
| 1905 |
+
"loss": 4.7596,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6007730535615682,
|
| 1910 |
+
"grad_norm": 7.693032264709473,
|
| 1911 |
+
"learning_rate": 5.178461669194903e-05,
|
| 1912 |
+
"loss": 4.7534,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6029817780231916,
|
| 1917 |
+
"grad_norm": 8.456910133361816,
|
| 1918 |
+
"learning_rate": 5.133858691783123e-05,
|
| 1919 |
+
"loss": 4.9262,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.605190502484815,
|
| 1924 |
+
"grad_norm": 8.400936126708984,
|
| 1925 |
+
"learning_rate": 5.08924505189423e-05,
|
| 1926 |
+
"loss": 4.4905,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.6073992269464384,
|
| 1931 |
+
"grad_norm": 8.744009971618652,
|
| 1932 |
+
"learning_rate": 5.0446243032153875e-05,
|
| 1933 |
+
"loss": 4.5625,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.6096079514080619,
|
| 1938 |
+
"grad_norm": 8.910327911376953,
|
| 1939 |
+
"learning_rate": 5e-05,
|
| 1940 |
+
"loss": 5.2096,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.6118166758696852,
|
| 1945 |
+
"grad_norm": 8.787812232971191,
|
| 1946 |
+
"learning_rate": 4.9553756967846144e-05,
|
| 1947 |
+
"loss": 4.8109,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.6140254003313087,
|
| 1952 |
+
"grad_norm": 8.353130340576172,
|
| 1953 |
+
"learning_rate": 4.9107549481057696e-05,
|
| 1954 |
+
"loss": 4.6905,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.616234124792932,
|
| 1959 |
+
"grad_norm": 8.456555366516113,
|
| 1960 |
+
"learning_rate": 4.866141308216878e-05,
|
| 1961 |
+
"loss": 5.3734,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.6184428492545555,
|
| 1966 |
+
"grad_norm": 8.801351547241211,
|
| 1967 |
+
"learning_rate": 4.821538330805098e-05,
|
| 1968 |
+
"loss": 5.1219,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.620651573716179,
|
| 1973 |
+
"grad_norm": 8.711759567260742,
|
| 1974 |
+
"learning_rate": 4.7769495687082725e-05,
|
| 1975 |
+
"loss": 5.0409,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.6228602981778023,
|
| 1980 |
+
"grad_norm": 8.708940505981445,
|
| 1981 |
+
"learning_rate": 4.7323785736319244e-05,
|
| 1982 |
+
"loss": 4.6526,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.6250690226394258,
|
| 1987 |
+
"grad_norm": 8.809032440185547,
|
| 1988 |
+
"learning_rate": 4.687828895866355e-05,
|
| 1989 |
+
"loss": 4.8223,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.6272777471010491,
|
| 1994 |
+
"grad_norm": 9.175921440124512,
|
| 1995 |
+
"learning_rate": 4.643304084003839e-05,
|
| 1996 |
+
"loss": 4.8745,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.6294864715626726,
|
| 2001 |
+
"grad_norm": 8.915994644165039,
|
| 2002 |
+
"learning_rate": 4.598807684655963e-05,
|
| 2003 |
+
"loss": 5.0015,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.631695196024296,
|
| 2008 |
+
"grad_norm": 9.178176879882812,
|
| 2009 |
+
"learning_rate": 4.5543432421711205e-05,
|
| 2010 |
+
"loss": 4.9042,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.6339039204859194,
|
| 2015 |
+
"grad_norm": 8.78802490234375,
|
| 2016 |
+
"learning_rate": 4.509914298352197e-05,
|
| 2017 |
+
"loss": 5.0455,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.6361126449475428,
|
| 2022 |
+
"grad_norm": 9.149216651916504,
|
| 2023 |
+
"learning_rate": 4.4655243921744374e-05,
|
| 2024 |
+
"loss": 4.8976,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.6383213694091662,
|
| 2029 |
+
"grad_norm": 9.336906433105469,
|
| 2030 |
+
"learning_rate": 4.4211770595035496e-05,
|
| 2031 |
+
"loss": 4.7158,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6405300938707896,
|
| 2036 |
+
"grad_norm": 8.577296257019043,
|
| 2037 |
+
"learning_rate": 4.3768758328140705e-05,
|
| 2038 |
+
"loss": 4.5859,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.6427388183324131,
|
| 2043 |
+
"grad_norm": 9.4404878616333,
|
| 2044 |
+
"learning_rate": 4.3326242409079705e-05,
|
| 2045 |
+
"loss": 4.8293,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.6449475427940364,
|
| 2050 |
+
"grad_norm": 9.113490104675293,
|
| 2051 |
+
"learning_rate": 4.288425808633575e-05,
|
| 2052 |
+
"loss": 4.9918,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.6471562672556599,
|
| 2057 |
+
"grad_norm": 9.25007438659668,
|
| 2058 |
+
"learning_rate": 4.2442840566047964e-05,
|
| 2059 |
+
"loss": 4.7895,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.6493649917172832,
|
| 2064 |
+
"grad_norm": 9.28573226928711,
|
| 2065 |
+
"learning_rate": 4.2002025009206994e-05,
|
| 2066 |
+
"loss": 4.7842,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.6515737161789067,
|
| 2071 |
+
"grad_norm": 9.722284317016602,
|
| 2072 |
+
"learning_rate": 4.1561846528854214e-05,
|
| 2073 |
+
"loss": 4.6345,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.6537824406405301,
|
| 2078 |
+
"grad_norm": 9.87505054473877,
|
| 2079 |
+
"learning_rate": 4.1122340187284846e-05,
|
| 2080 |
+
"loss": 4.5099,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.6559911651021535,
|
| 2085 |
+
"grad_norm": 10.381636619567871,
|
| 2086 |
+
"learning_rate": 4.068354099325511e-05,
|
| 2087 |
+
"loss": 4.8878,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.658199889563777,
|
| 2092 |
+
"grad_norm": 10.358430862426758,
|
| 2093 |
+
"learning_rate": 4.0245483899193595e-05,
|
| 2094 |
+
"loss": 4.7117,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.6604086140254003,
|
| 2099 |
+
"grad_norm": 10.86618423461914,
|
| 2100 |
+
"learning_rate": 3.980820379841708e-05,
|
| 2101 |
+
"loss": 4.8503,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.6626173384870238,
|
| 2106 |
+
"grad_norm": 11.921982765197754,
|
| 2107 |
+
"learning_rate": 3.937173552235117e-05,
|
| 2108 |
+
"loss": 5.1041,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.6648260629486472,
|
| 2113 |
+
"grad_norm": 4.309291362762451,
|
| 2114 |
+
"learning_rate": 3.893611383775582e-05,
|
| 2115 |
+
"loss": 3.8454,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.6670347874102706,
|
| 2120 |
+
"grad_norm": 5.9008097648620605,
|
| 2121 |
+
"learning_rate": 3.850137344395598e-05,
|
| 2122 |
+
"loss": 4.865,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.669243511871894,
|
| 2127 |
+
"grad_norm": 5.888951778411865,
|
| 2128 |
+
"learning_rate": 3.806754897007759e-05,
|
| 2129 |
+
"loss": 4.3102,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.6714522363335174,
|
| 2134 |
+
"grad_norm": 6.510290145874023,
|
| 2135 |
+
"learning_rate": 3.763467497228922e-05,
|
| 2136 |
+
"loss": 4.4619,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.6736609607951408,
|
| 2141 |
+
"grad_norm": 6.538812637329102,
|
| 2142 |
+
"learning_rate": 3.720278593104963e-05,
|
| 2143 |
+
"loss": 4.5339,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.6758696852567642,
|
| 2148 |
+
"grad_norm": 6.5220136642456055,
|
| 2149 |
+
"learning_rate": 3.677191624836106e-05,
|
| 2150 |
+
"loss": 4.7157,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.6780784097183876,
|
| 2155 |
+
"grad_norm": 6.203037261962891,
|
| 2156 |
+
"learning_rate": 3.634210024502903e-05,
|
| 2157 |
+
"loss": 4.4569,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.6802871341800111,
|
| 2162 |
+
"grad_norm": 6.234178066253662,
|
| 2163 |
+
"learning_rate": 3.591337215792852e-05,
|
| 2164 |
+
"loss": 4.6308,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.6824958586416344,
|
| 2169 |
+
"grad_norm": 6.102400302886963,
|
| 2170 |
+
"learning_rate": 3.5485766137276894e-05,
|
| 2171 |
+
"loss": 4.6082,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.6847045831032579,
|
| 2176 |
+
"grad_norm": 6.148480415344238,
|
| 2177 |
+
"learning_rate": 3.5059316243913554e-05,
|
| 2178 |
+
"loss": 4.6146,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.6869133075648812,
|
| 2183 |
+
"grad_norm": 5.826088905334473,
|
| 2184 |
+
"learning_rate": 3.463405644658695e-05,
|
| 2185 |
+
"loss": 4.6873,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.6891220320265047,
|
| 2190 |
+
"grad_norm": 5.779773235321045,
|
| 2191 |
+
"learning_rate": 3.421002061924876e-05,
|
| 2192 |
+
"loss": 4.452,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.6913307564881281,
|
| 2197 |
+
"grad_norm": 5.726971626281738,
|
| 2198 |
+
"learning_rate": 3.378724253835564e-05,
|
| 2199 |
+
"loss": 4.0362,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.6935394809497515,
|
| 2204 |
+
"grad_norm": 5.8791422843933105,
|
| 2205 |
+
"learning_rate": 3.336575588017881e-05,
|
| 2206 |
+
"loss": 4.5656,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.6957482054113749,
|
| 2211 |
+
"grad_norm": 5.984518527984619,
|
| 2212 |
+
"learning_rate": 3.294559421812158e-05,
|
| 2213 |
+
"loss": 4.2669,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.6979569298729983,
|
| 2218 |
+
"grad_norm": 6.276485919952393,
|
| 2219 |
+
"learning_rate": 3.2526791020045086e-05,
|
| 2220 |
+
"loss": 4.521,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.7001656543346217,
|
| 2225 |
+
"grad_norm": 6.756010055541992,
|
| 2226 |
+
"learning_rate": 3.210937964560234e-05,
|
| 2227 |
+
"loss": 4.2707,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.7023743787962452,
|
| 2232 |
+
"grad_norm": 6.564029216766357,
|
| 2233 |
+
"learning_rate": 3.1693393343581044e-05,
|
| 2234 |
+
"loss": 4.0805,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.7045831032578685,
|
| 2239 |
+
"grad_norm": 6.940414905548096,
|
| 2240 |
+
"learning_rate": 3.127886524925515e-05,
|
| 2241 |
+
"loss": 4.4171,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.706791827719492,
|
| 2246 |
+
"grad_norm": 7.548460006713867,
|
| 2247 |
+
"learning_rate": 3.086582838174551e-05,
|
| 2248 |
+
"loss": 4.8604,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.7090005521811154,
|
| 2253 |
+
"grad_norm": 7.360044002532959,
|
| 2254 |
+
"learning_rate": 3.045431564138969e-05,
|
| 2255 |
+
"loss": 4.6096,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.7112092766427388,
|
| 2260 |
+
"grad_norm": 8.025870323181152,
|
| 2261 |
+
"learning_rate": 3.004435980712129e-05,
|
| 2262 |
+
"loss": 4.4699,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.7134180011043623,
|
| 2267 |
+
"grad_norm": 7.835132598876953,
|
| 2268 |
+
"learning_rate": 2.9635993533859062e-05,
|
| 2269 |
+
"loss": 4.7898,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.7156267255659856,
|
| 2274 |
+
"grad_norm": 7.707442760467529,
|
| 2275 |
+
"learning_rate": 2.9229249349905684e-05,
|
| 2276 |
+
"loss": 4.6073,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.7178354500276091,
|
| 2281 |
+
"grad_norm": 7.932573318481445,
|
| 2282 |
+
"learning_rate": 2.8824159654356732e-05,
|
| 2283 |
+
"loss": 4.0933,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.7200441744892324,
|
| 2288 |
+
"grad_norm": 8.290323257446289,
|
| 2289 |
+
"learning_rate": 2.842075671451996e-05,
|
| 2290 |
+
"loss": 4.4136,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.7222528989508559,
|
| 2295 |
+
"grad_norm": 8.424840927124023,
|
| 2296 |
+
"learning_rate": 2.801907266334516e-05,
|
| 2297 |
+
"loss": 4.5087,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.7244616234124793,
|
| 2302 |
+
"grad_norm": 8.157886505126953,
|
| 2303 |
+
"learning_rate": 2.7619139496864378e-05,
|
| 2304 |
+
"loss": 4.3176,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.7266703478741027,
|
| 2309 |
+
"grad_norm": 8.439778327941895,
|
| 2310 |
+
"learning_rate": 2.7220989071643533e-05,
|
| 2311 |
+
"loss": 4.386,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.7288790723357261,
|
| 2316 |
+
"grad_norm": 8.862435340881348,
|
| 2317 |
+
"learning_rate": 2.682465310224473e-05,
|
| 2318 |
+
"loss": 4.922,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7310877967973495,
|
| 2323 |
+
"grad_norm": 8.862676620483398,
|
| 2324 |
+
"learning_rate": 2.6430163158700115e-05,
|
| 2325 |
+
"loss": 4.3482,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.7332965212589729,
|
| 2330 |
+
"grad_norm": 8.734305381774902,
|
| 2331 |
+
"learning_rate": 2.603755066399718e-05,
|
| 2332 |
+
"loss": 4.5575,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.7355052457205964,
|
| 2337 |
+
"grad_norm": 8.912976264953613,
|
| 2338 |
+
"learning_rate": 2.5646846891575637e-05,
|
| 2339 |
+
"loss": 4.3409,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.7377139701822197,
|
| 2344 |
+
"grad_norm": 8.8455810546875,
|
| 2345 |
+
"learning_rate": 2.5258082962836614e-05,
|
| 2346 |
+
"loss": 4.3128,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.7399226946438432,
|
| 2351 |
+
"grad_norm": 9.515101432800293,
|
| 2352 |
+
"learning_rate": 2.4871289844663437e-05,
|
| 2353 |
+
"loss": 4.9582,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.7421314191054665,
|
| 2358 |
+
"grad_norm": 9.195880889892578,
|
| 2359 |
+
"learning_rate": 2.4486498346955027e-05,
|
| 2360 |
+
"loss": 4.9631,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.74434014356709,
|
| 2365 |
+
"grad_norm": 8.75934886932373,
|
| 2366 |
+
"learning_rate": 2.410373912017185e-05,
|
| 2367 |
+
"loss": 4.5848,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.7465488680287135,
|
| 2372 |
+
"grad_norm": 9.588723182678223,
|
| 2373 |
+
"learning_rate": 2.3723042652894362e-05,
|
| 2374 |
+
"loss": 4.3137,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.7487575924903368,
|
| 2379 |
+
"grad_norm": 9.760534286499023,
|
| 2380 |
+
"learning_rate": 2.334443926939448e-05,
|
| 2381 |
+
"loss": 4.7105,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.7509663169519603,
|
| 2386 |
+
"grad_norm": 9.867559432983398,
|
| 2387 |
+
"learning_rate": 2.296795912722014e-05,
|
| 2388 |
+
"loss": 4.7552,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.7531750414135836,
|
| 2393 |
+
"grad_norm": 10.123164176940918,
|
| 2394 |
+
"learning_rate": 2.2593632214792994e-05,
|
| 2395 |
+
"loss": 4.7314,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.7553837658752071,
|
| 2400 |
+
"grad_norm": 10.126588821411133,
|
| 2401 |
+
"learning_rate": 2.2221488349019903e-05,
|
| 2402 |
+
"loss": 4.284,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.7575924903368305,
|
| 2407 |
+
"grad_norm": 10.301033973693848,
|
| 2408 |
+
"learning_rate": 2.1851557172917624e-05,
|
| 2409 |
+
"loss": 4.4204,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.7598012147984539,
|
| 2414 |
+
"grad_norm": 9.673257827758789,
|
| 2415 |
+
"learning_rate": 2.148386815325179e-05,
|
| 2416 |
+
"loss": 4.5415,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.7620099392600773,
|
| 2421 |
+
"grad_norm": 10.736601829528809,
|
| 2422 |
+
"learning_rate": 2.111845057818972e-05,
|
| 2423 |
+
"loss": 4.5998,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.7642186637217007,
|
| 2428 |
+
"grad_norm": 10.613544464111328,
|
| 2429 |
+
"learning_rate": 2.075533355496735e-05,
|
| 2430 |
+
"loss": 4.9721,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.7664273881833241,
|
| 2435 |
+
"grad_norm": 10.81152629852295,
|
| 2436 |
+
"learning_rate": 2.0394546007570857e-05,
|
| 2437 |
+
"loss": 4.6632,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.7686361126449476,
|
| 2442 |
+
"grad_norm": 11.234309196472168,
|
| 2443 |
+
"learning_rate": 2.0036116674432654e-05,
|
| 2444 |
+
"loss": 4.6279,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.7708448371065709,
|
| 2449 |
+
"grad_norm": 11.452018737792969,
|
| 2450 |
+
"learning_rate": 1.9680074106142238e-05,
|
| 2451 |
+
"loss": 4.8806,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.7730535615681944,
|
| 2456 |
+
"grad_norm": 13.051356315612793,
|
| 2457 |
+
"learning_rate": 1.9326446663172036e-05,
|
| 2458 |
+
"loss": 4.6126,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.7752622860298177,
|
| 2463 |
+
"grad_norm": 3.898439884185791,
|
| 2464 |
+
"learning_rate": 1.8975262513618293e-05,
|
| 2465 |
+
"loss": 3.3373,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.7774710104914412,
|
| 2470 |
+
"grad_norm": 5.114429473876953,
|
| 2471 |
+
"learning_rate": 1.8626549630957396e-05,
|
| 2472 |
+
"loss": 4.1765,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.7796797349530646,
|
| 2477 |
+
"grad_norm": 5.5542497634887695,
|
| 2478 |
+
"learning_rate": 1.8280335791817733e-05,
|
| 2479 |
+
"loss": 4.5676,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.781888459414688,
|
| 2484 |
+
"grad_norm": 5.338506698608398,
|
| 2485 |
+
"learning_rate": 1.7936648573766957e-05,
|
| 2486 |
+
"loss": 4.2831,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.7840971838763114,
|
| 2491 |
+
"grad_norm": 6.327260971069336,
|
| 2492 |
+
"learning_rate": 1.759551535311548e-05,
|
| 2493 |
+
"loss": 4.238,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.7863059083379348,
|
| 2498 |
+
"grad_norm": 5.520748138427734,
|
| 2499 |
+
"learning_rate": 1.725696330273575e-05,
|
| 2500 |
+
"loss": 4.406,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.7885146327995582,
|
| 2505 |
+
"grad_norm": 5.61647891998291,
|
| 2506 |
+
"learning_rate": 1.6921019389897795e-05,
|
| 2507 |
+
"loss": 4.5205,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.7907233572611817,
|
| 2512 |
+
"grad_norm": 6.23562479019165,
|
| 2513 |
+
"learning_rate": 1.6587710374121203e-05,
|
| 2514 |
+
"loss": 4.5157,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.792932081722805,
|
| 2519 |
+
"grad_norm": 5.914584159851074,
|
| 2520 |
+
"learning_rate": 1.6257062805043478e-05,
|
| 2521 |
+
"loss": 4.5219,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.7951408061844285,
|
| 2526 |
+
"grad_norm": 5.929191589355469,
|
| 2527 |
+
"learning_rate": 1.592910302030544e-05,
|
| 2528 |
+
"loss": 4.3534,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.7973495306460519,
|
| 2533 |
+
"grad_norm": 5.869836330413818,
|
| 2534 |
+
"learning_rate": 1.5603857143453128e-05,
|
| 2535 |
+
"loss": 4.2266,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.7995582551076753,
|
| 2540 |
+
"grad_norm": 6.144409656524658,
|
| 2541 |
+
"learning_rate": 1.5281351081856974e-05,
|
| 2542 |
+
"loss": 4.194,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.8017669795692988,
|
| 2547 |
+
"grad_norm": 6.3378520011901855,
|
| 2548 |
+
"learning_rate": 1.4961610524648223e-05,
|
| 2549 |
+
"loss": 4.8193,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.8039757040309221,
|
| 2554 |
+
"grad_norm": 6.32310152053833,
|
| 2555 |
+
"learning_rate": 1.4644660940672627e-05,
|
| 2556 |
+
"loss": 4.4801,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8061844284925456,
|
| 2561 |
+
"grad_norm": 6.283787250518799,
|
| 2562 |
+
"learning_rate": 1.4330527576461728e-05,
|
| 2563 |
+
"loss": 4.4156,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.8083931529541689,
|
| 2568 |
+
"grad_norm": 6.670344352722168,
|
| 2569 |
+
"learning_rate": 1.4019235454221857e-05,
|
| 2570 |
+
"loss": 4.0886,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.8106018774157924,
|
| 2575 |
+
"grad_norm": 6.906120300292969,
|
| 2576 |
+
"learning_rate": 1.371080936984101e-05,
|
| 2577 |
+
"loss": 4.4984,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.8128106018774158,
|
| 2582 |
+
"grad_norm": 7.01516580581665,
|
| 2583 |
+
"learning_rate": 1.340527389091374e-05,
|
| 2584 |
+
"loss": 4.2786,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.8150193263390392,
|
| 2589 |
+
"grad_norm": 7.165297508239746,
|
| 2590 |
+
"learning_rate": 1.3102653354784177e-05,
|
| 2591 |
+
"loss": 4.765,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.8172280508006626,
|
| 2596 |
+
"grad_norm": 7.451290130615234,
|
| 2597 |
+
"learning_rate": 1.2802971866607521e-05,
|
| 2598 |
+
"loss": 4.4154,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.819436775262286,
|
| 2603 |
+
"grad_norm": 7.890896320343018,
|
| 2604 |
+
"learning_rate": 1.250625329742996e-05,
|
| 2605 |
+
"loss": 4.6839,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.8216454997239094,
|
| 2610 |
+
"grad_norm": 8.337573051452637,
|
| 2611 |
+
"learning_rate": 1.2212521282287092e-05,
|
| 2612 |
+
"loss": 4.0614,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.8238542241855329,
|
| 2617 |
+
"grad_norm": 7.795760154724121,
|
| 2618 |
+
"learning_rate": 1.192179921832145e-05,
|
| 2619 |
+
"loss": 4.3992,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.8260629486471562,
|
| 2624 |
+
"grad_norm": 7.948381423950195,
|
| 2625 |
+
"learning_rate": 1.1634110262918718e-05,
|
| 2626 |
+
"loss": 4.4898,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8282716731087797,
|
| 2631 |
+
"grad_norm": 7.649637222290039,
|
| 2632 |
+
"learning_rate": 1.134947733186315e-05,
|
| 2633 |
+
"loss": 4.6223,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.830480397570403,
|
| 2638 |
+
"grad_norm": 8.012591361999512,
|
| 2639 |
+
"learning_rate": 1.1067923097512256e-05,
|
| 2640 |
+
"loss": 4.696,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.8326891220320265,
|
| 2645 |
+
"grad_norm": 8.13010025024414,
|
| 2646 |
+
"learning_rate": 1.078946998699073e-05,
|
| 2647 |
+
"loss": 4.3658,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.83489784649365,
|
| 2652 |
+
"grad_norm": 8.306947708129883,
|
| 2653 |
+
"learning_rate": 1.0514140180404204e-05,
|
| 2654 |
+
"loss": 4.9003,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.8371065709552733,
|
| 2659 |
+
"grad_norm": 8.446629524230957,
|
| 2660 |
+
"learning_rate": 1.0241955609072358e-05,
|
| 2661 |
+
"loss": 4.9169,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.8393152954168968,
|
| 2666 |
+
"grad_norm": 8.176511764526367,
|
| 2667 |
+
"learning_rate": 9.972937953781986e-06,
|
| 2668 |
+
"loss": 4.1928,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.8415240198785201,
|
| 2673 |
+
"grad_norm": 8.192519187927246,
|
| 2674 |
+
"learning_rate": 9.707108643060115e-06,
|
| 2675 |
+
"loss": 4.0346,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.8437327443401436,
|
| 2680 |
+
"grad_norm": 8.429787635803223,
|
| 2681 |
+
"learning_rate": 9.444488851467042e-06,
|
| 2682 |
+
"loss": 4.3153,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.845941468801767,
|
| 2687 |
+
"grad_norm": 8.412949562072754,
|
| 2688 |
+
"learning_rate": 9.185099497909716e-06,
|
| 2689 |
+
"loss": 4.5692,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.8481501932633904,
|
| 2694 |
+
"grad_norm": 8.446182250976562,
|
| 2695 |
+
"learning_rate": 8.928961243975437e-06,
|
| 2696 |
+
"loss": 4.6245,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.8503589177250138,
|
| 2701 |
+
"grad_norm": 9.340173721313477,
|
| 2702 |
+
"learning_rate": 8.676094492286013e-06,
|
| 2703 |
+
"loss": 4.5607,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.8525676421866372,
|
| 2708 |
+
"grad_norm": 8.916312217712402,
|
| 2709 |
+
"learning_rate": 8.426519384872733e-06,
|
| 2710 |
+
"loss": 4.6569,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.8547763666482606,
|
| 2715 |
+
"grad_norm": 8.926466941833496,
|
| 2716 |
+
"learning_rate": 8.180255801571824e-06,
|
| 2717 |
+
"loss": 4.8406,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.8569850911098841,
|
| 2722 |
+
"grad_norm": 8.984560012817383,
|
| 2723 |
+
"learning_rate": 7.937323358440935e-06,
|
| 2724 |
+
"loss": 4.5565,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.8591938155715074,
|
| 2729 |
+
"grad_norm": 9.738987922668457,
|
| 2730 |
+
"learning_rate": 7.697741406196757e-06,
|
| 2731 |
+
"loss": 4.5395,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.8614025400331309,
|
| 2736 |
+
"grad_norm": 9.285807609558105,
|
| 2737 |
+
"learning_rate": 7.461529028673464e-06,
|
| 2738 |
+
"loss": 4.595,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.8636112644947542,
|
| 2743 |
+
"grad_norm": 9.441668510437012,
|
| 2744 |
+
"learning_rate": 7.228705041302741e-06,
|
| 2745 |
+
"loss": 4.5406,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.8658199889563777,
|
| 2750 |
+
"grad_norm": 9.5773344039917,
|
| 2751 |
+
"learning_rate": 6.999287989614972e-06,
|
| 2752 |
+
"loss": 3.9756,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.8680287134180011,
|
| 2757 |
+
"grad_norm": 9.876778602600098,
|
| 2758 |
+
"learning_rate": 6.773296147762037e-06,
|
| 2759 |
+
"loss": 4.7135,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.8702374378796245,
|
| 2764 |
+
"grad_norm": 10.206405639648438,
|
| 2765 |
+
"learning_rate": 6.5507475170616565e-06,
|
| 2766 |
+
"loss": 4.7689,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.8724461623412479,
|
| 2771 |
+
"grad_norm": 9.651083946228027,
|
| 2772 |
+
"learning_rate": 6.331659824563513e-06,
|
| 2773 |
+
"loss": 4.3547,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.8746548868028713,
|
| 2778 |
+
"grad_norm": 9.921658515930176,
|
| 2779 |
+
"learning_rate": 6.116050521637218e-06,
|
| 2780 |
+
"loss": 4.5168,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.8768636112644947,
|
| 2785 |
+
"grad_norm": 11.268795013427734,
|
| 2786 |
+
"learning_rate": 5.903936782582253e-06,
|
| 2787 |
+
"loss": 4.4979,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.8790723357261182,
|
| 2792 |
+
"grad_norm": 10.472535133361816,
|
| 2793 |
+
"learning_rate": 5.69533550325988e-06,
|
| 2794 |
+
"loss": 4.4414,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.8812810601877415,
|
| 2799 |
+
"grad_norm": 11.554786682128906,
|
| 2800 |
+
"learning_rate": 5.4902632997473665e-06,
|
| 2801 |
+
"loss": 4.6124,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.883489784649365,
|
| 2806 |
+
"grad_norm": 13.761305809020996,
|
| 2807 |
+
"learning_rate": 5.288736507014435e-06,
|
| 2808 |
+
"loss": 5.3981,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.8856985091109884,
|
| 2813 |
+
"grad_norm": 4.258291244506836,
|
| 2814 |
+
"learning_rate": 5.090771177622083e-06,
|
| 2815 |
+
"loss": 3.1499,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.8879072335726118,
|
| 2820 |
+
"grad_norm": 4.202602386474609,
|
| 2821 |
+
"learning_rate": 4.896383080443934e-06,
|
| 2822 |
+
"loss": 3.9556,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.8901159580342353,
|
| 2827 |
+
"grad_norm": 5.0006513595581055,
|
| 2828 |
+
"learning_rate": 4.705587699410136e-06,
|
| 2829 |
+
"loss": 4.307,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.8923246824958586,
|
| 2834 |
+
"grad_norm": 4.940323829650879,
|
| 2835 |
+
"learning_rate": 4.5184002322740785e-06,
|
| 2836 |
+
"loss": 4.4528,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.8945334069574821,
|
| 2841 |
+
"grad_norm": 5.05031681060791,
|
| 2842 |
+
"learning_rate": 4.3348355894017424e-06,
|
| 2843 |
+
"loss": 4.0433,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.8967421314191054,
|
| 2848 |
+
"grad_norm": 5.438938617706299,
|
| 2849 |
+
"learning_rate": 4.154908392584017e-06,
|
| 2850 |
+
"loss": 4.2909,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.8989508558807289,
|
| 2855 |
+
"grad_norm": 5.099344253540039,
|
| 2856 |
+
"learning_rate": 3.978632973872065e-06,
|
| 2857 |
+
"loss": 4.1406,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.9011595803423523,
|
| 2862 |
+
"grad_norm": 5.504106521606445,
|
| 2863 |
+
"learning_rate": 3.8060233744356633e-06,
|
| 2864 |
+
"loss": 4.3558,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.9033683048039757,
|
| 2869 |
+
"grad_norm": 5.649275302886963,
|
| 2870 |
+
"learning_rate": 3.6370933434447698e-06,
|
| 2871 |
+
"loss": 4.4172,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.9055770292655991,
|
| 2876 |
+
"grad_norm": 5.748392105102539,
|
| 2877 |
+
"learning_rate": 3.471856336974322e-06,
|
| 2878 |
+
"loss": 4.1043,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.9077857537272225,
|
| 2883 |
+
"grad_norm": 5.7138237953186035,
|
| 2884 |
+
"learning_rate": 3.310325516932422e-06,
|
| 2885 |
+
"loss": 4.5152,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.9099944781888459,
|
| 2890 |
+
"grad_norm": 5.844823837280273,
|
| 2891 |
+
"learning_rate": 3.1525137500119207e-06,
|
| 2892 |
+
"loss": 4.3754,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.9122032026504694,
|
| 2897 |
+
"grad_norm": 5.811467170715332,
|
| 2898 |
+
"learning_rate": 2.998433606665496e-06,
|
| 2899 |
+
"loss": 4.4152,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.9144119271120927,
|
| 2904 |
+
"grad_norm": 5.769085884094238,
|
| 2905 |
+
"learning_rate": 2.848097360104396e-06,
|
| 2906 |
+
"loss": 4.4248,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.9166206515737162,
|
| 2911 |
+
"grad_norm": 6.123106479644775,
|
| 2912 |
+
"learning_rate": 2.7015169853208235e-06,
|
| 2913 |
+
"loss": 4.4335,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.9188293760353395,
|
| 2918 |
+
"grad_norm": 6.143897533416748,
|
| 2919 |
+
"learning_rate": 2.5587041581340233e-06,
|
| 2920 |
+
"loss": 4.4991,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.921038100496963,
|
| 2925 |
+
"grad_norm": 6.4361677169799805,
|
| 2926 |
+
"learning_rate": 2.419670254260281e-06,
|
| 2927 |
+
"loss": 4.3968,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.9232468249585865,
|
| 2932 |
+
"grad_norm": 6.608926773071289,
|
| 2933 |
+
"learning_rate": 2.2844263484068096e-06,
|
| 2934 |
+
"loss": 4.4498,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.9254555494202098,
|
| 2939 |
+
"grad_norm": 7.144963264465332,
|
| 2940 |
+
"learning_rate": 2.152983213389559e-06,
|
| 2941 |
+
"loss": 4.6328,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.9276642738818333,
|
| 2946 |
+
"grad_norm": 7.143801212310791,
|
| 2947 |
+
"learning_rate": 2.0253513192751373e-06,
|
| 2948 |
+
"loss": 4.4744,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.9298729983434566,
|
| 2953 |
+
"grad_norm": 7.323301315307617,
|
| 2954 |
+
"learning_rate": 1.9015408325467888e-06,
|
| 2955 |
+
"loss": 4.4851,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.9320817228050801,
|
| 2960 |
+
"grad_norm": 7.94163179397583,
|
| 2961 |
+
"learning_rate": 1.7815616152946523e-06,
|
| 2962 |
+
"loss": 4.705,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.9342904472667035,
|
| 2967 |
+
"grad_norm": 7.78623628616333,
|
| 2968 |
+
"learning_rate": 1.6654232244301294e-06,
|
| 2969 |
+
"loss": 4.4493,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.9364991717283269,
|
| 2974 |
+
"grad_norm": 7.809777736663818,
|
| 2975 |
+
"learning_rate": 1.5531349109246362e-06,
|
| 2976 |
+
"loss": 4.6519,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.9387078961899503,
|
| 2981 |
+
"grad_norm": 7.850180149078369,
|
| 2982 |
+
"learning_rate": 1.4447056190727725e-06,
|
| 2983 |
+
"loss": 4.3355,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.9409166206515737,
|
| 2988 |
+
"grad_norm": 8.725530624389648,
|
| 2989 |
+
"learning_rate": 1.3401439857798292e-06,
|
| 2990 |
+
"loss": 4.4763,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.9431253451131971,
|
| 2995 |
+
"grad_norm": 7.9435834884643555,
|
| 2996 |
+
"learning_rate": 1.2394583398738114e-06,
|
| 2997 |
+
"loss": 4.0902,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.9453340695748206,
|
| 3002 |
+
"grad_norm": 8.458505630493164,
|
| 3003 |
+
"learning_rate": 1.1426567014420297e-06,
|
| 3004 |
+
"loss": 4.4556,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.9475427940364439,
|
| 3009 |
+
"grad_norm": 8.368391036987305,
|
| 3010 |
+
"learning_rate": 1.0497467811922346e-06,
|
| 3011 |
+
"loss": 4.6413,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.9497515184980674,
|
| 3016 |
+
"grad_norm": 8.806220054626465,
|
| 3017 |
+
"learning_rate": 9.607359798384785e-07,
|
| 3018 |
+
"loss": 4.2521,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.9519602429596907,
|
| 3023 |
+
"grad_norm": 8.404871940612793,
|
| 3024 |
+
"learning_rate": 8.756313875115385e-07,
|
| 3025 |
+
"loss": 4.2418,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.9541689674213142,
|
| 3030 |
+
"grad_norm": 8.731966018676758,
|
| 3031 |
+
"learning_rate": 7.944397831941952e-07,
|
| 3032 |
+
"loss": 4.5369,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.9563776918829376,
|
| 3037 |
+
"grad_norm": 8.71829605102539,
|
| 3038 |
+
"learning_rate": 7.171676341812706e-07,
|
| 3039 |
+
"loss": 4.7399,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.958586416344561,
|
| 3044 |
+
"grad_norm": 8.327555656433105,
|
| 3045 |
+
"learning_rate": 6.438210955644452e-07,
|
| 3046 |
+
"loss": 4.5096,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.9607951408061844,
|
| 3051 |
+
"grad_norm": 8.708925247192383,
|
| 3052 |
+
"learning_rate": 5.74406009741979e-07,
|
| 3053 |
+
"loss": 4.5694,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.9630038652678078,
|
| 3058 |
+
"grad_norm": 9.103087425231934,
|
| 3059 |
+
"learning_rate": 5.089279059533658e-07,
|
| 3060 |
+
"loss": 4.4516,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.9652125897294312,
|
| 3065 |
+
"grad_norm": 8.804319381713867,
|
| 3066 |
+
"learning_rate": 4.4739199983886425e-07,
|
| 3067 |
+
"loss": 4.4854,
|
| 3068 |
+
"step": 437
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.9674213141910547,
|
| 3072 |
+
"grad_norm": 8.990171432495117,
|
| 3073 |
+
"learning_rate": 3.8980319302407977e-07,
|
| 3074 |
+
"loss": 4.7322,
|
| 3075 |
+
"step": 438
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.969630038652678,
|
| 3079 |
+
"grad_norm": 9.266417503356934,
|
| 3080 |
+
"learning_rate": 3.361660727295046e-07,
|
| 3081 |
+
"loss": 4.8561,
|
| 3082 |
+
"step": 439
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.9718387631143015,
|
| 3086 |
+
"grad_norm": 9.650453567504883,
|
| 3087 |
+
"learning_rate": 2.8648491140513266e-07,
|
| 3088 |
+
"loss": 5.1407,
|
| 3089 |
+
"step": 440
|
| 3090 |
+
},
|
| 3091 |
+
{
|
| 3092 |
+
"epoch": 0.9740474875759249,
|
| 3093 |
+
"grad_norm": 9.014724731445312,
|
| 3094 |
+
"learning_rate": 2.407636663901591e-07,
|
| 3095 |
+
"loss": 4.7725,
|
| 3096 |
+
"step": 441
|
| 3097 |
+
},
|
| 3098 |
+
{
|
| 3099 |
+
"epoch": 0.9762562120375483,
|
| 3100 |
+
"grad_norm": 9.816120147705078,
|
| 3101 |
+
"learning_rate": 1.9900597959770507e-07,
|
| 3102 |
+
"loss": 4.3926,
|
| 3103 |
+
"step": 442
|
| 3104 |
+
},
|
| 3105 |
+
{
|
| 3106 |
+
"epoch": 0.9784649364991718,
|
| 3107 |
+
"grad_norm": 9.809698104858398,
|
| 3108 |
+
"learning_rate": 1.612151772247883e-07,
|
| 3109 |
+
"loss": 4.1767,
|
| 3110 |
+
"step": 443
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"epoch": 0.9806736609607951,
|
| 3114 |
+
"grad_norm": 9.744946479797363,
|
| 3115 |
+
"learning_rate": 1.2739426948732424e-07,
|
| 3116 |
+
"loss": 4.2795,
|
| 3117 |
+
"step": 444
|
| 3118 |
+
},
|
| 3119 |
+
{
|
| 3120 |
+
"epoch": 0.9828823854224186,
|
| 3121 |
+
"grad_norm": 9.93597412109375,
|
| 3122 |
+
"learning_rate": 9.754595038037328e-08,
|
| 3123 |
+
"loss": 4.1773,
|
| 3124 |
+
"step": 445
|
| 3125 |
+
},
|
| 3126 |
+
{
|
| 3127 |
+
"epoch": 0.9850911098840419,
|
| 3128 |
+
"grad_norm": 10.203986167907715,
|
| 3129 |
+
"learning_rate": 7.167259746355681e-08,
|
| 3130 |
+
"loss": 4.3431,
|
| 3131 |
+
"step": 446
|
| 3132 |
+
},
|
| 3133 |
+
{
|
| 3134 |
+
"epoch": 0.9872998343456654,
|
| 3135 |
+
"grad_norm": 9.65555191040039,
|
| 3136 |
+
"learning_rate": 4.9776271671642115e-08,
|
| 3137 |
+
"loss": 4.5657,
|
| 3138 |
+
"step": 447
|
| 3139 |
+
},
|
| 3140 |
+
{
|
| 3141 |
+
"epoch": 0.9895085588072888,
|
| 3142 |
+
"grad_norm": 11.336668014526367,
|
| 3143 |
+
"learning_rate": 3.185871715041255e-08,
|
| 3144 |
+
"loss": 5.1661,
|
| 3145 |
+
"step": 448
|
| 3146 |
+
},
|
| 3147 |
+
{
|
| 3148 |
+
"epoch": 0.9917172832689122,
|
| 3149 |
+
"grad_norm": 11.222773551940918,
|
| 3150 |
+
"learning_rate": 1.7921361117734236e-08,
|
| 3151 |
+
"loss": 4.8351,
|
| 3152 |
+
"step": 449
|
| 3153 |
+
},
|
| 3154 |
+
{
|
| 3155 |
+
"epoch": 0.9939260077305356,
|
| 3156 |
+
"grad_norm": 13.293135643005371,
|
| 3157 |
+
"learning_rate": 7.96531374983589e-09,
|
| 3158 |
+
"loss": 4.995,
|
| 3159 |
+
"step": 450
|
| 3160 |
+
},
|
| 3161 |
+
{
|
| 3162 |
+
"epoch": 0.996134732192159,
|
| 3163 |
+
"grad_norm": 4.857739448547363,
|
| 3164 |
+
"learning_rate": 1.9913680929295465e-09,
|
| 3165 |
+
"loss": 3.7627,
|
| 3166 |
+
"step": 451
|
| 3167 |
+
},
|
| 3168 |
+
{
|
| 3169 |
+
"epoch": 0.9983434566537824,
|
| 3170 |
+
"grad_norm": 8.479840278625488,
|
| 3171 |
+
"learning_rate": 0.0,
|
| 3172 |
+
"loss": 4.7239,
|
| 3173 |
+
"step": 452
|
| 3174 |
}
|
| 3175 |
],
|
| 3176 |
"logging_steps": 1,
|
|
|
|
| 3185 |
"should_evaluate": false,
|
| 3186 |
"should_log": false,
|
| 3187 |
"should_save": true,
|
| 3188 |
+
"should_training_stop": true
|
| 3189 |
},
|
| 3190 |
"attributes": {}
|
| 3191 |
}
|
| 3192 |
},
|
| 3193 |
+
"total_flos": 1.7098768232585626e+17,
|
| 3194 |
"train_batch_size": 4,
|
| 3195 |
"trial_name": null,
|
| 3196 |
"trial_params": null
|