Training in progress, step 356, 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 125918320
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e481995cf6d237020e6aab4a1e7404637d3ca4faf64b643a18771e95eaa340eb
|
| 3 |
size 125918320
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c5f87aedc82c3a0d97bdd0013b3b34f2f23143184a093d55e9a14b33b4ccd6b
|
| 3 |
+
size 64684244
|
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 sha256:2f940370db0768ac9a318b750a653da4911f1db917a73143506c211e803ef55e
|
| 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 sha256:17d88a80230fc1c998d77e01d3856d0912d9f20dd4f54f9e81578aede7b8a3ec
|
| 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,
|
|
@@ -1253,6 +1253,1252 @@
|
|
| 1253 |
"learning_rate": 8.590166950903118e-05,
|
| 1254 |
"loss": 0.6124,
|
| 1255 |
"step": 178
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1256 |
}
|
| 1257 |
],
|
| 1258 |
"logging_steps": 1,
|
|
@@ -1272,7 +2518,7 @@
|
|
| 1272 |
"attributes": {}
|
| 1273 |
}
|
| 1274 |
},
|
| 1275 |
-
"total_flos":
|
| 1276 |
"train_batch_size": 4,
|
| 1277 |
"trial_name": null,
|
| 1278 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5012319605772615,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 356,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1253 |
"learning_rate": 8.590166950903118e-05,
|
| 1254 |
"loss": 0.6124,
|
| 1255 |
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.2520239352340725,
|
| 1259 |
+
"grad_norm": 14.029097557067871,
|
| 1260 |
+
"learning_rate": 8.574645793874449e-05,
|
| 1261 |
+
"loss": 0.8471,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.25343189017951423,
|
| 1266 |
+
"grad_norm": 34.442237854003906,
|
| 1267 |
+
"learning_rate": 8.559053854876494e-05,
|
| 1268 |
+
"loss": 1.8921,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.254839845124956,
|
| 1273 |
+
"grad_norm": 17.151165008544922,
|
| 1274 |
+
"learning_rate": 8.543391442647013e-05,
|
| 1275 |
+
"loss": 1.1393,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.25624780007039777,
|
| 1280 |
+
"grad_norm": 16.576353073120117,
|
| 1281 |
+
"learning_rate": 8.527658867319221e-05,
|
| 1282 |
+
"loss": 0.4902,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.2576557550158395,
|
| 1287 |
+
"grad_norm": 31.439992904663086,
|
| 1288 |
+
"learning_rate": 8.511856440415635e-05,
|
| 1289 |
+
"loss": 1.7562,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.25906370996128125,
|
| 1294 |
+
"grad_norm": 102.67780303955078,
|
| 1295 |
+
"learning_rate": 8.495984474841918e-05,
|
| 1296 |
+
"loss": 4.32,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.26047166490672297,
|
| 1301 |
+
"grad_norm": 45.334354400634766,
|
| 1302 |
+
"learning_rate": 8.480043284880666e-05,
|
| 1303 |
+
"loss": 2.0922,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.26187961985216474,
|
| 1308 |
+
"grad_norm": 37.032569885253906,
|
| 1309 |
+
"learning_rate": 8.464033186185203e-05,
|
| 1310 |
+
"loss": 1.9404,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.26328757479760645,
|
| 1315 |
+
"grad_norm": 33.529842376708984,
|
| 1316 |
+
"learning_rate": 8.44795449577332e-05,
|
| 1317 |
+
"loss": 1.5912,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.2646955297430482,
|
| 1322 |
+
"grad_norm": 53.00758361816406,
|
| 1323 |
+
"learning_rate": 8.431807532021e-05,
|
| 1324 |
+
"loss": 3.0904,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.26610348468849,
|
| 1329 |
+
"grad_norm": 31.37575340270996,
|
| 1330 |
+
"learning_rate": 8.415592614656119e-05,
|
| 1331 |
+
"loss": 1.6186,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.2675114396339317,
|
| 1336 |
+
"grad_norm": 29.855907440185547,
|
| 1337 |
+
"learning_rate": 8.399310064752099e-05,
|
| 1338 |
+
"loss": 1.8013,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.26891939457937347,
|
| 1343 |
+
"grad_norm": 43.18759536743164,
|
| 1344 |
+
"learning_rate": 8.382960204721575e-05,
|
| 1345 |
+
"loss": 1.6222,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.2703273495248152,
|
| 1350 |
+
"grad_norm": 31.552494049072266,
|
| 1351 |
+
"learning_rate": 8.366543358309989e-05,
|
| 1352 |
+
"loss": 1.1728,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.27173530447025696,
|
| 1357 |
+
"grad_norm": 14.046395301818848,
|
| 1358 |
+
"learning_rate": 8.35005985058919e-05,
|
| 1359 |
+
"loss": 0.6825,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.2731432594156987,
|
| 1364 |
+
"grad_norm": 23.78717613220215,
|
| 1365 |
+
"learning_rate": 8.333510007950995e-05,
|
| 1366 |
+
"loss": 1.9934,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.27455121436114044,
|
| 1371 |
+
"grad_norm": 19.5039005279541,
|
| 1372 |
+
"learning_rate": 8.316894158100727e-05,
|
| 1373 |
+
"loss": 1.3594,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.2759591693065822,
|
| 1378 |
+
"grad_norm": 14.477341651916504,
|
| 1379 |
+
"learning_rate": 8.300212630050726e-05,
|
| 1380 |
+
"loss": 0.7189,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.2773671242520239,
|
| 1385 |
+
"grad_norm": 17.81009864807129,
|
| 1386 |
+
"learning_rate": 8.28346575411383e-05,
|
| 1387 |
+
"loss": 1.1535,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.2787750791974657,
|
| 1392 |
+
"grad_norm": 17.172264099121094,
|
| 1393 |
+
"learning_rate": 8.266653861896843e-05,
|
| 1394 |
+
"loss": 1.169,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.2801830341429074,
|
| 1399 |
+
"grad_norm": 57.817726135253906,
|
| 1400 |
+
"learning_rate": 8.24977728629396e-05,
|
| 1401 |
+
"loss": 2.9405,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.2815909890883492,
|
| 1406 |
+
"grad_norm": 177.61752319335938,
|
| 1407 |
+
"learning_rate": 8.232836361480183e-05,
|
| 1408 |
+
"loss": 4.7211,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.28299894403379094,
|
| 1413 |
+
"grad_norm": 27.973918914794922,
|
| 1414 |
+
"learning_rate": 8.215831422904694e-05,
|
| 1415 |
+
"loss": 1.0981,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.28440689897923266,
|
| 1420 |
+
"grad_norm": 29.390439987182617,
|
| 1421 |
+
"learning_rate": 8.198762807284224e-05,
|
| 1422 |
+
"loss": 1.7834,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.28581485392467443,
|
| 1427 |
+
"grad_norm": 35.740665435791016,
|
| 1428 |
+
"learning_rate": 8.181630852596379e-05,
|
| 1429 |
+
"loss": 2.2235,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.28722280887011614,
|
| 1434 |
+
"grad_norm": 26.734773635864258,
|
| 1435 |
+
"learning_rate": 8.164435898072947e-05,
|
| 1436 |
+
"loss": 1.9502,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.2886307638155579,
|
| 1441 |
+
"grad_norm": 18.85955047607422,
|
| 1442 |
+
"learning_rate": 8.147178284193184e-05,
|
| 1443 |
+
"loss": 1.2279,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.2900387187609996,
|
| 1448 |
+
"grad_norm": 17.08832550048828,
|
| 1449 |
+
"learning_rate": 8.129858352677076e-05,
|
| 1450 |
+
"loss": 1.1731,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.2914466737064414,
|
| 1455 |
+
"grad_norm": 44.938209533691406,
|
| 1456 |
+
"learning_rate": 8.112476446478561e-05,
|
| 1457 |
+
"loss": 1.5057,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.29285462865188316,
|
| 1462 |
+
"grad_norm": 43.36068344116211,
|
| 1463 |
+
"learning_rate": 8.095032909778748e-05,
|
| 1464 |
+
"loss": 1.0289,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.2942625835973249,
|
| 1469 |
+
"grad_norm": 18.839191436767578,
|
| 1470 |
+
"learning_rate": 8.077528087979095e-05,
|
| 1471 |
+
"loss": 1.1196,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.29567053854276665,
|
| 1476 |
+
"grad_norm": 20.551124572753906,
|
| 1477 |
+
"learning_rate": 8.05996232769458e-05,
|
| 1478 |
+
"loss": 1.1649,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.29707849348820836,
|
| 1483 |
+
"grad_norm": 25.39565658569336,
|
| 1484 |
+
"learning_rate": 8.042335976746822e-05,
|
| 1485 |
+
"loss": 1.3211,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.29848644843365013,
|
| 1490 |
+
"grad_norm": 14.85946273803711,
|
| 1491 |
+
"learning_rate": 8.024649384157212e-05,
|
| 1492 |
+
"loss": 0.6612,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.29989440337909185,
|
| 1497 |
+
"grad_norm": 19.78711700439453,
|
| 1498 |
+
"learning_rate": 8.006902900139988e-05,
|
| 1499 |
+
"loss": 1.3747,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.3013023583245336,
|
| 1504 |
+
"grad_norm": 11.568204879760742,
|
| 1505 |
+
"learning_rate": 7.989096876095303e-05,
|
| 1506 |
+
"loss": 1.1045,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.3027103132699754,
|
| 1511 |
+
"grad_norm": 9.248348236083984,
|
| 1512 |
+
"learning_rate": 7.971231664602273e-05,
|
| 1513 |
+
"loss": 0.7642,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.3041182682154171,
|
| 1518 |
+
"grad_norm": 5.907835960388184,
|
| 1519 |
+
"learning_rate": 7.953307619411987e-05,
|
| 1520 |
+
"loss": 0.3467,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.30552622316085887,
|
| 1525 |
+
"grad_norm": 7.299631118774414,
|
| 1526 |
+
"learning_rate": 7.935325095440511e-05,
|
| 1527 |
+
"loss": 0.3262,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.3069341781063006,
|
| 1532 |
+
"grad_norm": 11.987982749938965,
|
| 1533 |
+
"learning_rate": 7.917284448761854e-05,
|
| 1534 |
+
"loss": 0.6308,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.30834213305174235,
|
| 1539 |
+
"grad_norm": 10.527399063110352,
|
| 1540 |
+
"learning_rate": 7.899186036600918e-05,
|
| 1541 |
+
"loss": 1.0346,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.30975008799718406,
|
| 1546 |
+
"grad_norm": 45.371761322021484,
|
| 1547 |
+
"learning_rate": 7.88103021732643e-05,
|
| 1548 |
+
"loss": 2.1641,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.31115804294262583,
|
| 1553 |
+
"grad_norm": 19.207111358642578,
|
| 1554 |
+
"learning_rate": 7.862817350443839e-05,
|
| 1555 |
+
"loss": 1.1476,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.3125659978880676,
|
| 1560 |
+
"grad_norm": 59.24346160888672,
|
| 1561 |
+
"learning_rate": 7.8445477965882e-05,
|
| 1562 |
+
"loss": 1.891,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.3139739528335093,
|
| 1567 |
+
"grad_norm": 8.811090469360352,
|
| 1568 |
+
"learning_rate": 7.826221917517034e-05,
|
| 1569 |
+
"loss": 0.5215,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.3153819077789511,
|
| 1574 |
+
"grad_norm": 33.62078094482422,
|
| 1575 |
+
"learning_rate": 7.807840076103163e-05,
|
| 1576 |
+
"loss": 1.5825,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.3167898627243928,
|
| 1581 |
+
"grad_norm": 31.49905014038086,
|
| 1582 |
+
"learning_rate": 7.789402636327525e-05,
|
| 1583 |
+
"loss": 1.084,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.31819781766983457,
|
| 1588 |
+
"grad_norm": 11.637606620788574,
|
| 1589 |
+
"learning_rate": 7.770909963271972e-05,
|
| 1590 |
+
"loss": 0.7103,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.3196057726152763,
|
| 1595 |
+
"grad_norm": 21.97195816040039,
|
| 1596 |
+
"learning_rate": 7.752362423112032e-05,
|
| 1597 |
+
"loss": 1.3204,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.32101372756071805,
|
| 1602 |
+
"grad_norm": 18.698328018188477,
|
| 1603 |
+
"learning_rate": 7.733760383109665e-05,
|
| 1604 |
+
"loss": 0.9943,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.3224216825061598,
|
| 1609 |
+
"grad_norm": 19.418983459472656,
|
| 1610 |
+
"learning_rate": 7.715104211605987e-05,
|
| 1611 |
+
"loss": 0.8113,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.32382963745160154,
|
| 1616 |
+
"grad_norm": 16.439781188964844,
|
| 1617 |
+
"learning_rate": 7.696394278013979e-05,
|
| 1618 |
+
"loss": 0.6862,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.3252375923970433,
|
| 1623 |
+
"grad_norm": 9.741683959960938,
|
| 1624 |
+
"learning_rate": 7.677630952811172e-05,
|
| 1625 |
+
"loss": 0.5196,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.326645547342485,
|
| 1630 |
+
"grad_norm": 8.357406616210938,
|
| 1631 |
+
"learning_rate": 7.65881460753231e-05,
|
| 1632 |
+
"loss": 0.5758,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.3280535022879268,
|
| 1637 |
+
"grad_norm": 33.68247604370117,
|
| 1638 |
+
"learning_rate": 7.63994561476199e-05,
|
| 1639 |
+
"loss": 1.8643,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.32946145723336856,
|
| 1644 |
+
"grad_norm": 41.963356018066406,
|
| 1645 |
+
"learning_rate": 7.621024348127295e-05,
|
| 1646 |
+
"loss": 2.0034,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.3308694121788103,
|
| 1651 |
+
"grad_norm": 20.343597412109375,
|
| 1652 |
+
"learning_rate": 7.602051182290382e-05,
|
| 1653 |
+
"loss": 1.3085,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.33227736712425204,
|
| 1658 |
+
"grad_norm": 19.10648536682129,
|
| 1659 |
+
"learning_rate": 7.583026492941075e-05,
|
| 1660 |
+
"loss": 1.1332,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.33368532206969376,
|
| 1665 |
+
"grad_norm": 35.396820068359375,
|
| 1666 |
+
"learning_rate": 7.563950656789416e-05,
|
| 1667 |
+
"loss": 1.2122,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.3350932770151355,
|
| 1672 |
+
"grad_norm": 27.122194290161133,
|
| 1673 |
+
"learning_rate": 7.544824051558217e-05,
|
| 1674 |
+
"loss": 1.5366,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.33650123196057724,
|
| 1679 |
+
"grad_norm": 16.57398223876953,
|
| 1680 |
+
"learning_rate": 7.525647055975567e-05,
|
| 1681 |
+
"loss": 1.3327,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.337909186906019,
|
| 1686 |
+
"grad_norm": 28.07727813720703,
|
| 1687 |
+
"learning_rate": 7.506420049767347e-05,
|
| 1688 |
+
"loss": 1.3062,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.3393171418514608,
|
| 1693 |
+
"grad_norm": 24.03339385986328,
|
| 1694 |
+
"learning_rate": 7.4871434136497e-05,
|
| 1695 |
+
"loss": 1.6172,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.3407250967969025,
|
| 1700 |
+
"grad_norm": 12.594011306762695,
|
| 1701 |
+
"learning_rate": 7.467817529321498e-05,
|
| 1702 |
+
"loss": 1.0874,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.34213305174234426,
|
| 1707 |
+
"grad_norm": 12.581107139587402,
|
| 1708 |
+
"learning_rate": 7.448442779456781e-05,
|
| 1709 |
+
"loss": 0.7604,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.343541006687786,
|
| 1714 |
+
"grad_norm": 46.649383544921875,
|
| 1715 |
+
"learning_rate": 7.429019547697183e-05,
|
| 1716 |
+
"loss": 1.3192,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.34494896163322775,
|
| 1721 |
+
"grad_norm": 16.450733184814453,
|
| 1722 |
+
"learning_rate": 7.409548218644332e-05,
|
| 1723 |
+
"loss": 1.1468,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.34635691657866946,
|
| 1728 |
+
"grad_norm": 16.57231903076172,
|
| 1729 |
+
"learning_rate": 7.390029177852237e-05,
|
| 1730 |
+
"loss": 1.0065,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.34776487152411123,
|
| 1735 |
+
"grad_norm": 27.110193252563477,
|
| 1736 |
+
"learning_rate": 7.37046281181965e-05,
|
| 1737 |
+
"loss": 0.9427,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.349172826469553,
|
| 1742 |
+
"grad_norm": 17.80980110168457,
|
| 1743 |
+
"learning_rate": 7.35084950798242e-05,
|
| 1744 |
+
"loss": 1.184,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.3505807814149947,
|
| 1749 |
+
"grad_norm": 63.84309387207031,
|
| 1750 |
+
"learning_rate": 7.331189654705815e-05,
|
| 1751 |
+
"loss": 1.1323,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.3519887363604365,
|
| 1756 |
+
"grad_norm": 104.79566955566406,
|
| 1757 |
+
"learning_rate": 7.311483641276833e-05,
|
| 1758 |
+
"loss": 3.3279,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.3533966913058782,
|
| 1763 |
+
"grad_norm": 20.52196502685547,
|
| 1764 |
+
"learning_rate": 7.291731857896492e-05,
|
| 1765 |
+
"loss": 0.7327,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.35480464625131997,
|
| 1770 |
+
"grad_norm": 10.260120391845703,
|
| 1771 |
+
"learning_rate": 7.271934695672112e-05,
|
| 1772 |
+
"loss": 0.9386,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.3562126011967617,
|
| 1777 |
+
"grad_norm": 16.224411010742188,
|
| 1778 |
+
"learning_rate": 7.252092546609558e-05,
|
| 1779 |
+
"loss": 1.2396,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.35762055614220345,
|
| 1784 |
+
"grad_norm": 39.729496002197266,
|
| 1785 |
+
"learning_rate": 7.23220580360549e-05,
|
| 1786 |
+
"loss": 1.3439,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.3590285110876452,
|
| 1791 |
+
"grad_norm": 10.264772415161133,
|
| 1792 |
+
"learning_rate": 7.212274860439576e-05,
|
| 1793 |
+
"loss": 0.7011,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.36043646603308693,
|
| 1798 |
+
"grad_norm": 18.108182907104492,
|
| 1799 |
+
"learning_rate": 7.192300111766696e-05,
|
| 1800 |
+
"loss": 0.8619,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.3618444209785287,
|
| 1805 |
+
"grad_norm": 61.82547378540039,
|
| 1806 |
+
"learning_rate": 7.172281953109128e-05,
|
| 1807 |
+
"loss": 1.6629,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.3632523759239704,
|
| 1812 |
+
"grad_norm": 29.251495361328125,
|
| 1813 |
+
"learning_rate": 7.152220780848713e-05,
|
| 1814 |
+
"loss": 0.8569,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.3646603308694122,
|
| 1819 |
+
"grad_norm": 15.582707405090332,
|
| 1820 |
+
"learning_rate": 7.132116992219015e-05,
|
| 1821 |
+
"loss": 1.1466,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.3660682858148539,
|
| 1826 |
+
"grad_norm": 26.372283935546875,
|
| 1827 |
+
"learning_rate": 7.111970985297446e-05,
|
| 1828 |
+
"loss": 1.9285,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.36747624076029567,
|
| 1833 |
+
"grad_norm": 27.972610473632812,
|
| 1834 |
+
"learning_rate": 7.09178315899739e-05,
|
| 1835 |
+
"loss": 1.5448,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.36888419570573744,
|
| 1840 |
+
"grad_norm": 16.148784637451172,
|
| 1841 |
+
"learning_rate": 7.071553913060296e-05,
|
| 1842 |
+
"loss": 1.3117,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.37029215065117915,
|
| 1847 |
+
"grad_norm": 15.186225891113281,
|
| 1848 |
+
"learning_rate": 7.051283648047776e-05,
|
| 1849 |
+
"loss": 1.1989,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.3717001055966209,
|
| 1854 |
+
"grad_norm": 11.998665809631348,
|
| 1855 |
+
"learning_rate": 7.030972765333655e-05,
|
| 1856 |
+
"loss": 0.8488,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.37310806054206264,
|
| 1861 |
+
"grad_norm": 22.099727630615234,
|
| 1862 |
+
"learning_rate": 7.010621667096041e-05,
|
| 1863 |
+
"loss": 0.5446,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.3745160154875044,
|
| 1868 |
+
"grad_norm": 22.740089416503906,
|
| 1869 |
+
"learning_rate": 6.990230756309355e-05,
|
| 1870 |
+
"loss": 0.7579,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.3759239704329461,
|
| 1875 |
+
"grad_norm": 21.371496200561523,
|
| 1876 |
+
"learning_rate": 6.969800436736347e-05,
|
| 1877 |
+
"loss": 0.8681,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.3773319253783879,
|
| 1882 |
+
"grad_norm": 5.607769966125488,
|
| 1883 |
+
"learning_rate": 6.949331112920106e-05,
|
| 1884 |
+
"loss": 0.3161,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.37873988032382966,
|
| 1889 |
+
"grad_norm": 23.731332778930664,
|
| 1890 |
+
"learning_rate": 6.928823190176051e-05,
|
| 1891 |
+
"loss": 1.8792,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.3801478352692714,
|
| 1896 |
+
"grad_norm": 23.237396240234375,
|
| 1897 |
+
"learning_rate": 6.908277074583895e-05,
|
| 1898 |
+
"loss": 1.7937,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.38155579021471314,
|
| 1903 |
+
"grad_norm": 15.7110595703125,
|
| 1904 |
+
"learning_rate": 6.887693172979623e-05,
|
| 1905 |
+
"loss": 1.28,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.38296374516015486,
|
| 1910 |
+
"grad_norm": 10.618301391601562,
|
| 1911 |
+
"learning_rate": 6.867071892947417e-05,
|
| 1912 |
+
"loss": 0.7942,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.3843717001055966,
|
| 1917 |
+
"grad_norm": 4.260011196136475,
|
| 1918 |
+
"learning_rate": 6.846413642811598e-05,
|
| 1919 |
+
"loss": 0.6144,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.38577965505103834,
|
| 1924 |
+
"grad_norm": 16.570066452026367,
|
| 1925 |
+
"learning_rate": 6.825718831628532e-05,
|
| 1926 |
+
"loss": 1.0566,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.3871876099964801,
|
| 1931 |
+
"grad_norm": 11.564096450805664,
|
| 1932 |
+
"learning_rate": 6.80498786917854e-05,
|
| 1933 |
+
"loss": 0.7064,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.3885955649419219,
|
| 1938 |
+
"grad_norm": 7.003159523010254,
|
| 1939 |
+
"learning_rate": 6.784221165957771e-05,
|
| 1940 |
+
"loss": 0.5354,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.3900035198873636,
|
| 1945 |
+
"grad_norm": 14.146154403686523,
|
| 1946 |
+
"learning_rate": 6.763419133170089e-05,
|
| 1947 |
+
"loss": 0.6516,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.39141147483280536,
|
| 1952 |
+
"grad_norm": 5.380551338195801,
|
| 1953 |
+
"learning_rate": 6.742582182718915e-05,
|
| 1954 |
+
"loss": 0.4163,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.3928194297782471,
|
| 1959 |
+
"grad_norm": 4.118032455444336,
|
| 1960 |
+
"learning_rate": 6.721710727199087e-05,
|
| 1961 |
+
"loss": 0.4664,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.39422738472368885,
|
| 1966 |
+
"grad_norm": 10.134897232055664,
|
| 1967 |
+
"learning_rate": 6.700805179888675e-05,
|
| 1968 |
+
"loss": 0.7125,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.3956353396691306,
|
| 1973 |
+
"grad_norm": 23.42674446105957,
|
| 1974 |
+
"learning_rate": 6.679865954740808e-05,
|
| 1975 |
+
"loss": 1.5089,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.39704329461457233,
|
| 1980 |
+
"grad_norm": 11.591614723205566,
|
| 1981 |
+
"learning_rate": 6.658893466375474e-05,
|
| 1982 |
+
"loss": 0.7875,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.3984512495600141,
|
| 1987 |
+
"grad_norm": 14.444954872131348,
|
| 1988 |
+
"learning_rate": 6.637888130071304e-05,
|
| 1989 |
+
"loss": 0.9125,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.3998592045054558,
|
| 1994 |
+
"grad_norm": 14.189472198486328,
|
| 1995 |
+
"learning_rate": 6.616850361757364e-05,
|
| 1996 |
+
"loss": 1.0319,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.4012671594508976,
|
| 2001 |
+
"grad_norm": 36.17814636230469,
|
| 2002 |
+
"learning_rate": 6.595780578004901e-05,
|
| 2003 |
+
"loss": 1.2356,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.4026751143963393,
|
| 2008 |
+
"grad_norm": 45.84112548828125,
|
| 2009 |
+
"learning_rate": 6.574679196019112e-05,
|
| 2010 |
+
"loss": 1.6351,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.40408306934178106,
|
| 2015 |
+
"grad_norm": 29.742900848388672,
|
| 2016 |
+
"learning_rate": 6.553546633630865e-05,
|
| 2017 |
+
"loss": 0.9819,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.40549102428722283,
|
| 2022 |
+
"grad_norm": 24.932228088378906,
|
| 2023 |
+
"learning_rate": 6.532383309288443e-05,
|
| 2024 |
+
"loss": 0.9133,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.40689897923266455,
|
| 2029 |
+
"grad_norm": 42.26826477050781,
|
| 2030 |
+
"learning_rate": 6.511189642049244e-05,
|
| 2031 |
+
"loss": 1.1545,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.4083069341781063,
|
| 2036 |
+
"grad_norm": 18.30242156982422,
|
| 2037 |
+
"learning_rate": 6.489966051571492e-05,
|
| 2038 |
+
"loss": 1.4322,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.40971488912354803,
|
| 2043 |
+
"grad_norm": 55.628604888916016,
|
| 2044 |
+
"learning_rate": 6.468712958105926e-05,
|
| 2045 |
+
"loss": 2.6193,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.4111228440689898,
|
| 2050 |
+
"grad_norm": 150.41041564941406,
|
| 2051 |
+
"learning_rate": 6.447430782487473e-05,
|
| 2052 |
+
"loss": 2.5549,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.4125307990144315,
|
| 2057 |
+
"grad_norm": 37.905914306640625,
|
| 2058 |
+
"learning_rate": 6.426119946126921e-05,
|
| 2059 |
+
"loss": 1.6198,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.4139387539598733,
|
| 2064 |
+
"grad_norm": 28.293848037719727,
|
| 2065 |
+
"learning_rate": 6.404780871002575e-05,
|
| 2066 |
+
"loss": 0.9741,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.41534670890531505,
|
| 2071 |
+
"grad_norm": 25.466543197631836,
|
| 2072 |
+
"learning_rate": 6.383413979651894e-05,
|
| 2073 |
+
"loss": 1.0264,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.41675466385075677,
|
| 2078 |
+
"grad_norm": 16.458620071411133,
|
| 2079 |
+
"learning_rate": 6.362019695163132e-05,
|
| 2080 |
+
"loss": 0.9326,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.41816261879619854,
|
| 2085 |
+
"grad_norm": 9.713531494140625,
|
| 2086 |
+
"learning_rate": 6.340598441166958e-05,
|
| 2087 |
+
"loss": 0.5798,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.41957057374164025,
|
| 2092 |
+
"grad_norm": 50.1083869934082,
|
| 2093 |
+
"learning_rate": 6.31915064182807e-05,
|
| 2094 |
+
"loss": 1.43,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.420978528687082,
|
| 2099 |
+
"grad_norm": 40.683753967285156,
|
| 2100 |
+
"learning_rate": 6.297676721836784e-05,
|
| 2101 |
+
"loss": 1.4706,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.42238648363252373,
|
| 2106 |
+
"grad_norm": 43.749794006347656,
|
| 2107 |
+
"learning_rate": 6.276177106400647e-05,
|
| 2108 |
+
"loss": 3.0438,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.4237944385779655,
|
| 2113 |
+
"grad_norm": 8.859760284423828,
|
| 2114 |
+
"learning_rate": 6.254652221235992e-05,
|
| 2115 |
+
"loss": 0.2372,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.4252023935234073,
|
| 2120 |
+
"grad_norm": 6.741225719451904,
|
| 2121 |
+
"learning_rate": 6.23310249255953e-05,
|
| 2122 |
+
"loss": 0.5755,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.426610348468849,
|
| 2127 |
+
"grad_norm": 21.322010040283203,
|
| 2128 |
+
"learning_rate": 6.211528347079896e-05,
|
| 2129 |
+
"loss": 1.0998,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.42801830341429076,
|
| 2134 |
+
"grad_norm": 11.496692657470703,
|
| 2135 |
+
"learning_rate": 6.189930211989209e-05,
|
| 2136 |
+
"loss": 0.8235,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.42942625835973247,
|
| 2141 |
+
"grad_norm": 7.478254318237305,
|
| 2142 |
+
"learning_rate": 6.168308514954602e-05,
|
| 2143 |
+
"loss": 0.6988,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.43083421330517424,
|
| 2148 |
+
"grad_norm": 16.896244049072266,
|
| 2149 |
+
"learning_rate": 6.146663684109773e-05,
|
| 2150 |
+
"loss": 1.1964,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.43224216825061595,
|
| 2155 |
+
"grad_norm": 14.001093864440918,
|
| 2156 |
+
"learning_rate": 6.124996148046478e-05,
|
| 2157 |
+
"loss": 1.1556,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.4336501231960577,
|
| 2162 |
+
"grad_norm": 15.864665031433105,
|
| 2163 |
+
"learning_rate": 6.103306335806077e-05,
|
| 2164 |
+
"loss": 1.031,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.4350580781414995,
|
| 2169 |
+
"grad_norm": 17.03790855407715,
|
| 2170 |
+
"learning_rate": 6.0815946768710176e-05,
|
| 2171 |
+
"loss": 0.676,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.4364660330869412,
|
| 2176 |
+
"grad_norm": 12.771721839904785,
|
| 2177 |
+
"learning_rate": 6.0598616011563324e-05,
|
| 2178 |
+
"loss": 0.8617,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.437873988032383,
|
| 2183 |
+
"grad_norm": 9.611161231994629,
|
| 2184 |
+
"learning_rate": 6.038107539001139e-05,
|
| 2185 |
+
"loss": 0.7115,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.4392819429778247,
|
| 2190 |
+
"grad_norm": 23.515108108520508,
|
| 2191 |
+
"learning_rate": 6.016332921160099e-05,
|
| 2192 |
+
"loss": 1.2061,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.44068989792326646,
|
| 2197 |
+
"grad_norm": 10.848517417907715,
|
| 2198 |
+
"learning_rate": 5.99453817879491e-05,
|
| 2199 |
+
"loss": 0.5806,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.4420978528687082,
|
| 2204 |
+
"grad_norm": 9.37192153930664,
|
| 2205 |
+
"learning_rate": 5.972723743465749e-05,
|
| 2206 |
+
"loss": 0.4661,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.44350580781414994,
|
| 2211 |
+
"grad_norm": 19.835905075073242,
|
| 2212 |
+
"learning_rate": 5.9508900471227416e-05,
|
| 2213 |
+
"loss": 0.7906,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.4449137627595917,
|
| 2218 |
+
"grad_norm": 10.338900566101074,
|
| 2219 |
+
"learning_rate": 5.9290375220973985e-05,
|
| 2220 |
+
"loss": 0.5141,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.4463217177050334,
|
| 2225 |
+
"grad_norm": 7.031118869781494,
|
| 2226 |
+
"learning_rate": 5.907166601094063e-05,
|
| 2227 |
+
"loss": 0.6734,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.4477296726504752,
|
| 2232 |
+
"grad_norm": 12.128201484680176,
|
| 2233 |
+
"learning_rate": 5.885277717181338e-05,
|
| 2234 |
+
"loss": 0.808,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.4491376275959169,
|
| 2239 |
+
"grad_norm": 18.01626968383789,
|
| 2240 |
+
"learning_rate": 5.86337130378351e-05,
|
| 2241 |
+
"loss": 0.9725,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.4505455825413587,
|
| 2246 |
+
"grad_norm": 19.51142692565918,
|
| 2247 |
+
"learning_rate": 5.8414477946719725e-05,
|
| 2248 |
+
"loss": 1.0292,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.45195353748680045,
|
| 2253 |
+
"grad_norm": 13.940886497497559,
|
| 2254 |
+
"learning_rate": 5.8195076239566304e-05,
|
| 2255 |
+
"loss": 0.7187,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.45336149243224216,
|
| 2260 |
+
"grad_norm": 9.016057014465332,
|
| 2261 |
+
"learning_rate": 5.797551226077308e-05,
|
| 2262 |
+
"loss": 0.5587,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.45476944737768393,
|
| 2267 |
+
"grad_norm": 7.983224391937256,
|
| 2268 |
+
"learning_rate": 5.775579035795145e-05,
|
| 2269 |
+
"loss": 0.469,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.45617740232312565,
|
| 2274 |
+
"grad_norm": 11.286262512207031,
|
| 2275 |
+
"learning_rate": 5.753591488183988e-05,
|
| 2276 |
+
"loss": 0.7711,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.4575853572685674,
|
| 2281 |
+
"grad_norm": 28.39682960510254,
|
| 2282 |
+
"learning_rate": 5.731589018621777e-05,
|
| 2283 |
+
"loss": 1.1854,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.45899331221400913,
|
| 2288 |
+
"grad_norm": 11.899818420410156,
|
| 2289 |
+
"learning_rate": 5.709572062781924e-05,
|
| 2290 |
+
"loss": 0.7849,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.4604012671594509,
|
| 2295 |
+
"grad_norm": 6.919939994812012,
|
| 2296 |
+
"learning_rate": 5.68754105662468e-05,
|
| 2297 |
+
"loss": 0.5996,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.46180922210489267,
|
| 2302 |
+
"grad_norm": 7.084524154663086,
|
| 2303 |
+
"learning_rate": 5.665496436388515e-05,
|
| 2304 |
+
"loss": 0.3707,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.4632171770503344,
|
| 2309 |
+
"grad_norm": 3.3268368244171143,
|
| 2310 |
+
"learning_rate": 5.643438638581472e-05,
|
| 2311 |
+
"loss": 0.184,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.46462513199577615,
|
| 2316 |
+
"grad_norm": 9.73562240600586,
|
| 2317 |
+
"learning_rate": 5.621368099972519e-05,
|
| 2318 |
+
"loss": 0.6516,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.46603308694121787,
|
| 2323 |
+
"grad_norm": 18.59488296508789,
|
| 2324 |
+
"learning_rate": 5.599285257582911e-05,
|
| 2325 |
+
"loss": 1.3035,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.46744104188665964,
|
| 2330 |
+
"grad_norm": 6.1787543296813965,
|
| 2331 |
+
"learning_rate": 5.577190548677529e-05,
|
| 2332 |
+
"loss": 0.3772,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.46884899683210135,
|
| 2337 |
+
"grad_norm": 13.894497871398926,
|
| 2338 |
+
"learning_rate": 5.555084410756226e-05,
|
| 2339 |
+
"loss": 0.6582,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.4702569517775431,
|
| 2344 |
+
"grad_norm": 28.024072647094727,
|
| 2345 |
+
"learning_rate": 5.532967281545162e-05,
|
| 2346 |
+
"loss": 1.1448,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.4716649067229849,
|
| 2351 |
+
"grad_norm": 13.692731857299805,
|
| 2352 |
+
"learning_rate": 5.510839598988137e-05,
|
| 2353 |
+
"loss": 1.0288,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.4730728616684266,
|
| 2358 |
+
"grad_norm": 16.94679832458496,
|
| 2359 |
+
"learning_rate": 5.4887018012379164e-05,
|
| 2360 |
+
"loss": 0.8068,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.4744808166138684,
|
| 2365 |
+
"grad_norm": 46.81898880004883,
|
| 2366 |
+
"learning_rate": 5.466554326647564e-05,
|
| 2367 |
+
"loss": 1.5328,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.4758887715593101,
|
| 2372 |
+
"grad_norm": 23.71940803527832,
|
| 2373 |
+
"learning_rate": 5.4443976137617447e-05,
|
| 2374 |
+
"loss": 1.0762,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.47729672650475186,
|
| 2379 |
+
"grad_norm": 16.341949462890625,
|
| 2380 |
+
"learning_rate": 5.422232101308064e-05,
|
| 2381 |
+
"loss": 0.7343,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.47870468145019357,
|
| 2386 |
+
"grad_norm": 11.016429901123047,
|
| 2387 |
+
"learning_rate": 5.400058228188363e-05,
|
| 2388 |
+
"loss": 0.9309,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.48011263639563534,
|
| 2393 |
+
"grad_norm": 17.73030662536621,
|
| 2394 |
+
"learning_rate": 5.377876433470031e-05,
|
| 2395 |
+
"loss": 0.9843,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.4815205913410771,
|
| 2400 |
+
"grad_norm": 28.34784698486328,
|
| 2401 |
+
"learning_rate": 5.355687156377319e-05,
|
| 2402 |
+
"loss": 0.8469,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.4829285462865188,
|
| 2407 |
+
"grad_norm": 15.134780883789062,
|
| 2408 |
+
"learning_rate": 5.333490836282635e-05,
|
| 2409 |
+
"loss": 1.1461,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.4843365012319606,
|
| 2414 |
+
"grad_norm": 11.261271476745605,
|
| 2415 |
+
"learning_rate": 5.311287912697845e-05,
|
| 2416 |
+
"loss": 0.8169,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.4857444561774023,
|
| 2421 |
+
"grad_norm": 13.379313468933105,
|
| 2422 |
+
"learning_rate": 5.289078825265573e-05,
|
| 2423 |
+
"loss": 0.6616,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.4871524111228441,
|
| 2428 |
+
"grad_norm": 35.41781997680664,
|
| 2429 |
+
"learning_rate": 5.266864013750493e-05,
|
| 2430 |
+
"loss": 1.8836,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.4885603660682858,
|
| 2435 |
+
"grad_norm": 12.9522705078125,
|
| 2436 |
+
"learning_rate": 5.244643918030623e-05,
|
| 2437 |
+
"loss": 0.4605,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.48996832101372756,
|
| 2442 |
+
"grad_norm": 22.15134620666504,
|
| 2443 |
+
"learning_rate": 5.222418978088614e-05,
|
| 2444 |
+
"loss": 0.8405,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.49137627595916933,
|
| 2449 |
+
"grad_norm": 26.150135040283203,
|
| 2450 |
+
"learning_rate": 5.200189634003041e-05,
|
| 2451 |
+
"loss": 1.0932,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.49278423090461104,
|
| 2456 |
+
"grad_norm": 55.37350845336914,
|
| 2457 |
+
"learning_rate": 5.177956325939678e-05,
|
| 2458 |
+
"loss": 2.3808,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.4941921858500528,
|
| 2463 |
+
"grad_norm": 20.655296325683594,
|
| 2464 |
+
"learning_rate": 5.155719494142799e-05,
|
| 2465 |
+
"loss": 0.4508,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.4956001407954945,
|
| 2470 |
+
"grad_norm": 6.21237850189209,
|
| 2471 |
+
"learning_rate": 5.133479578926445e-05,
|
| 2472 |
+
"loss": 0.5059,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.4970080957409363,
|
| 2477 |
+
"grad_norm": 42.47819519042969,
|
| 2478 |
+
"learning_rate": 5.111237020665718e-05,
|
| 2479 |
+
"loss": 1.501,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.498416050686378,
|
| 2484 |
+
"grad_norm": 23.130695343017578,
|
| 2485 |
+
"learning_rate": 5.088992259788049e-05,
|
| 2486 |
+
"loss": 1.1172,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.4998240056318198,
|
| 2491 |
+
"grad_norm": 18.151142120361328,
|
| 2492 |
+
"learning_rate": 5.06674573676449e-05,
|
| 2493 |
+
"loss": 0.8362,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.5012319605772615,
|
| 2498 |
+
"grad_norm": 17.91101837158203,
|
| 2499 |
+
"learning_rate": 5.044497892100981e-05,
|
| 2500 |
+
"loss": 0.5968,
|
| 2501 |
+
"step": 356
|
| 2502 |
}
|
| 2503 |
],
|
| 2504 |
"logging_steps": 1,
|
|
|
|
| 2518 |
"attributes": {}
|
| 2519 |
}
|
| 2520 |
},
|
| 2521 |
+
"total_flos": 3.720766820701962e+17,
|
| 2522 |
"train_batch_size": 4,
|
| 2523 |
"trial_name": null,
|
| 2524 |
"trial_params": null
|