Training in progress, step 346, 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 125248064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c39b303b3b24c90608dda6b811a1313074653c2b41eefe49d8b132629403952
|
| 3 |
size 125248064
|
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 sha256:5dfc4521350c95d9abc2abce112bce9288c164b96a81482af74bacb192d8ba2b
|
| 3 |
+
size 64220436
|
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:d8eb5c3002d7e8ae1e2b9def5fa5c5531d299883fc64b5898be75ce56c2463ee
|
| 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 sha256:799bba65fc813f88f98a0005d4dd9bc4b536663da3ba1e7b7fc073c9e9774986
|
| 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": 173,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1234,6 +1234,1225 @@
|
|
| 1234 |
"eval_samples_per_second": 8.978,
|
| 1235 |
"eval_steps_per_second": 4.505,
|
| 1236 |
"step": 173
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1237 |
}
|
| 1238 |
],
|
| 1239 |
"logging_steps": 1,
|
|
@@ -1253,7 +2472,7 @@
|
|
| 1253 |
"attributes": {}
|
| 1254 |
}
|
| 1255 |
},
|
| 1256 |
-
"total_flos":
|
| 1257 |
"train_batch_size": 2,
|
| 1258 |
"trial_name": null,
|
| 1259 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.502177068214804,
|
| 5 |
"eval_steps": 173,
|
| 6 |
+
"global_step": 346,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1234 |
"eval_samples_per_second": 8.978,
|
| 1235 |
"eval_steps_per_second": 4.505,
|
| 1236 |
"step": 173
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 0.2525399129172714,
|
| 1240 |
+
"grad_norm": 0.24334195256233215,
|
| 1241 |
+
"learning_rate": 0.00017256662525779983,
|
| 1242 |
+
"loss": 0.5441,
|
| 1243 |
+
"step": 174
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 0.2539912917271408,
|
| 1247 |
+
"grad_norm": 0.3937956988811493,
|
| 1248 |
+
"learning_rate": 0.00017224750460699556,
|
| 1249 |
+
"loss": 1.3447,
|
| 1250 |
+
"step": 175
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.25544267053701014,
|
| 1254 |
+
"grad_norm": 0.32860031723976135,
|
| 1255 |
+
"learning_rate": 0.00017192683734099353,
|
| 1256 |
+
"loss": 1.1908,
|
| 1257 |
+
"step": 176
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.25689404934687954,
|
| 1261 |
+
"grad_norm": 0.40750887989997864,
|
| 1262 |
+
"learning_rate": 0.000171604630324375,
|
| 1263 |
+
"loss": 0.4587,
|
| 1264 |
+
"step": 177
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.25834542815674894,
|
| 1268 |
+
"grad_norm": 0.24471426010131836,
|
| 1269 |
+
"learning_rate": 0.00017128089045468294,
|
| 1270 |
+
"loss": 0.4731,
|
| 1271 |
+
"step": 178
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 0.2597968069666183,
|
| 1275 |
+
"grad_norm": 0.3871125876903534,
|
| 1276 |
+
"learning_rate": 0.0001709556246622744,
|
| 1277 |
+
"loss": 1.5461,
|
| 1278 |
+
"step": 179
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 0.2612481857764877,
|
| 1282 |
+
"grad_norm": 0.32278481125831604,
|
| 1283 |
+
"learning_rate": 0.00017062883991017218,
|
| 1284 |
+
"loss": 1.0052,
|
| 1285 |
+
"step": 180
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.262699564586357,
|
| 1289 |
+
"grad_norm": 0.2906551957130432,
|
| 1290 |
+
"learning_rate": 0.00017030054319391562,
|
| 1291 |
+
"loss": 1.1581,
|
| 1292 |
+
"step": 181
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.2641509433962264,
|
| 1296 |
+
"grad_norm": 0.2662639319896698,
|
| 1297 |
+
"learning_rate": 0.00016997074154141096,
|
| 1298 |
+
"loss": 0.4337,
|
| 1299 |
+
"step": 182
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.2656023222060958,
|
| 1303 |
+
"grad_norm": 0.30343329906463623,
|
| 1304 |
+
"learning_rate": 0.00016963944201278097,
|
| 1305 |
+
"loss": 0.9429,
|
| 1306 |
+
"step": 183
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.26705370101596515,
|
| 1310 |
+
"grad_norm": 0.48314595222473145,
|
| 1311 |
+
"learning_rate": 0.00016930665170021363,
|
| 1312 |
+
"loss": 1.146,
|
| 1313 |
+
"step": 184
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.26850507982583455,
|
| 1317 |
+
"grad_norm": 0.22615785896778107,
|
| 1318 |
+
"learning_rate": 0.00016897237772781044,
|
| 1319 |
+
"loss": 0.288,
|
| 1320 |
+
"step": 185
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 0.26995645863570394,
|
| 1324 |
+
"grad_norm": 0.3443828225135803,
|
| 1325 |
+
"learning_rate": 0.00016863662725143382,
|
| 1326 |
+
"loss": 1.0044,
|
| 1327 |
+
"step": 186
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 0.2714078374455733,
|
| 1331 |
+
"grad_norm": 0.3521503508090973,
|
| 1332 |
+
"learning_rate": 0.00016829940745855408,
|
| 1333 |
+
"loss": 0.917,
|
| 1334 |
+
"step": 187
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.2728592162554427,
|
| 1338 |
+
"grad_norm": 0.3083937168121338,
|
| 1339 |
+
"learning_rate": 0.00016796072556809534,
|
| 1340 |
+
"loss": 1.1641,
|
| 1341 |
+
"step": 188
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.274310595065312,
|
| 1345 |
+
"grad_norm": 0.1671004444360733,
|
| 1346 |
+
"learning_rate": 0.0001676205888302812,
|
| 1347 |
+
"loss": 0.2087,
|
| 1348 |
+
"step": 189
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 0.2757619738751814,
|
| 1352 |
+
"grad_norm": 0.3343052268028259,
|
| 1353 |
+
"learning_rate": 0.00016727900452647926,
|
| 1354 |
+
"loss": 0.8929,
|
| 1355 |
+
"step": 190
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 0.2772133526850508,
|
| 1359 |
+
"grad_norm": 0.2790353000164032,
|
| 1360 |
+
"learning_rate": 0.00016693597996904564,
|
| 1361 |
+
"loss": 0.8818,
|
| 1362 |
+
"step": 191
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 0.27866473149492016,
|
| 1366 |
+
"grad_norm": 0.3753064274787903,
|
| 1367 |
+
"learning_rate": 0.00016659152250116812,
|
| 1368 |
+
"loss": 0.7531,
|
| 1369 |
+
"step": 192
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.28011611030478956,
|
| 1373 |
+
"grad_norm": 0.3089276850223541,
|
| 1374 |
+
"learning_rate": 0.00016624563949670904,
|
| 1375 |
+
"loss": 1.2895,
|
| 1376 |
+
"step": 193
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.28156748911465895,
|
| 1380 |
+
"grad_norm": 0.24410511553287506,
|
| 1381 |
+
"learning_rate": 0.00016589833836004752,
|
| 1382 |
+
"loss": 0.444,
|
| 1383 |
+
"step": 194
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 0.2830188679245283,
|
| 1387 |
+
"grad_norm": 0.4272465705871582,
|
| 1388 |
+
"learning_rate": 0.00016554962652592077,
|
| 1389 |
+
"loss": 0.9711,
|
| 1390 |
+
"step": 195
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.2844702467343977,
|
| 1394 |
+
"grad_norm": 0.29573267698287964,
|
| 1395 |
+
"learning_rate": 0.00016519951145926515,
|
| 1396 |
+
"loss": 0.3583,
|
| 1397 |
+
"step": 196
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 0.28592162554426703,
|
| 1401 |
+
"grad_norm": 0.42986953258514404,
|
| 1402 |
+
"learning_rate": 0.00016484800065505627,
|
| 1403 |
+
"loss": 1.6862,
|
| 1404 |
+
"step": 197
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.28737300435413643,
|
| 1408 |
+
"grad_norm": 0.31086039543151855,
|
| 1409 |
+
"learning_rate": 0.00016449510163814854,
|
| 1410 |
+
"loss": 0.5808,
|
| 1411 |
+
"step": 198
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 0.2888243831640058,
|
| 1415 |
+
"grad_norm": 0.4590548574924469,
|
| 1416 |
+
"learning_rate": 0.000164140821963114,
|
| 1417 |
+
"loss": 1.837,
|
| 1418 |
+
"step": 199
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.29027576197387517,
|
| 1422 |
+
"grad_norm": 0.4661678969860077,
|
| 1423 |
+
"learning_rate": 0.00016378516921408077,
|
| 1424 |
+
"loss": 0.8179,
|
| 1425 |
+
"step": 200
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.29172714078374457,
|
| 1429 |
+
"grad_norm": 0.5419331192970276,
|
| 1430 |
+
"learning_rate": 0.00016342815100457063,
|
| 1431 |
+
"loss": 0.8166,
|
| 1432 |
+
"step": 201
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 0.2931785195936139,
|
| 1436 |
+
"grad_norm": 0.34207096695899963,
|
| 1437 |
+
"learning_rate": 0.00016306977497733592,
|
| 1438 |
+
"loss": 0.8683,
|
| 1439 |
+
"step": 202
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.2946298984034833,
|
| 1443 |
+
"grad_norm": 0.4035762548446655,
|
| 1444 |
+
"learning_rate": 0.00016271004880419608,
|
| 1445 |
+
"loss": 0.7494,
|
| 1446 |
+
"step": 203
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.2960812772133527,
|
| 1450 |
+
"grad_norm": 0.40903428196907043,
|
| 1451 |
+
"learning_rate": 0.00016234898018587337,
|
| 1452 |
+
"loss": 0.3219,
|
| 1453 |
+
"step": 204
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.29753265602322204,
|
| 1457 |
+
"grad_norm": 0.3800163269042969,
|
| 1458 |
+
"learning_rate": 0.000161986576851828,
|
| 1459 |
+
"loss": 1.0624,
|
| 1460 |
+
"step": 205
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.29898403483309144,
|
| 1464 |
+
"grad_norm": 0.20540378987789154,
|
| 1465 |
+
"learning_rate": 0.00016162284656009274,
|
| 1466 |
+
"loss": 0.2874,
|
| 1467 |
+
"step": 206
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.30043541364296084,
|
| 1471 |
+
"grad_norm": 0.3501240313053131,
|
| 1472 |
+
"learning_rate": 0.00016125779709710665,
|
| 1473 |
+
"loss": 0.9726,
|
| 1474 |
+
"step": 207
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.3018867924528302,
|
| 1478 |
+
"grad_norm": 0.46438068151474,
|
| 1479 |
+
"learning_rate": 0.00016089143627754862,
|
| 1480 |
+
"loss": 0.9383,
|
| 1481 |
+
"step": 208
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.3033381712626996,
|
| 1485 |
+
"grad_norm": 0.37778106331825256,
|
| 1486 |
+
"learning_rate": 0.00016052377194417,
|
| 1487 |
+
"loss": 1.9048,
|
| 1488 |
+
"step": 209
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.3047895500725689,
|
| 1492 |
+
"grad_norm": 0.4382157325744629,
|
| 1493 |
+
"learning_rate": 0.0001601548119676266,
|
| 1494 |
+
"loss": 1.2582,
|
| 1495 |
+
"step": 210
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.3062409288824383,
|
| 1499 |
+
"grad_norm": 0.442871630191803,
|
| 1500 |
+
"learning_rate": 0.00015978456424631032,
|
| 1501 |
+
"loss": 1.086,
|
| 1502 |
+
"step": 211
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.3076923076923077,
|
| 1506 |
+
"grad_norm": 0.5801809430122375,
|
| 1507 |
+
"learning_rate": 0.00015941303670618018,
|
| 1508 |
+
"loss": 1.234,
|
| 1509 |
+
"step": 212
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.30914368650217705,
|
| 1513 |
+
"grad_norm": 0.2682696282863617,
|
| 1514 |
+
"learning_rate": 0.00015904023730059228,
|
| 1515 |
+
"loss": 0.6148,
|
| 1516 |
+
"step": 213
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 0.31059506531204645,
|
| 1520 |
+
"grad_norm": 0.25142902135849,
|
| 1521 |
+
"learning_rate": 0.00015866617401012994,
|
| 1522 |
+
"loss": 0.6304,
|
| 1523 |
+
"step": 214
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 0.31204644412191584,
|
| 1527 |
+
"grad_norm": 0.36662545800209045,
|
| 1528 |
+
"learning_rate": 0.00015829085484243266,
|
| 1529 |
+
"loss": 0.8005,
|
| 1530 |
+
"step": 215
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.3134978229317852,
|
| 1534 |
+
"grad_norm": 0.2825576961040497,
|
| 1535 |
+
"learning_rate": 0.00015791428783202465,
|
| 1536 |
+
"loss": 0.7511,
|
| 1537 |
+
"step": 216
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.3149492017416546,
|
| 1541 |
+
"grad_norm": 0.33648526668548584,
|
| 1542 |
+
"learning_rate": 0.00015753648104014297,
|
| 1543 |
+
"loss": 0.9242,
|
| 1544 |
+
"step": 217
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.3164005805515239,
|
| 1548 |
+
"grad_norm": 0.5301381945610046,
|
| 1549 |
+
"learning_rate": 0.00015715744255456496,
|
| 1550 |
+
"loss": 0.9624,
|
| 1551 |
+
"step": 218
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 0.3178519593613933,
|
| 1555 |
+
"grad_norm": 0.3330354690551758,
|
| 1556 |
+
"learning_rate": 0.00015677718048943496,
|
| 1557 |
+
"loss": 1.6149,
|
| 1558 |
+
"step": 219
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 0.3193033381712627,
|
| 1562 |
+
"grad_norm": 0.2863297164440155,
|
| 1563 |
+
"learning_rate": 0.00015639570298509064,
|
| 1564 |
+
"loss": 0.5041,
|
| 1565 |
+
"step": 220
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 0.32075471698113206,
|
| 1569 |
+
"grad_norm": 0.30266642570495605,
|
| 1570 |
+
"learning_rate": 0.00015601301820788898,
|
| 1571 |
+
"loss": 0.5615,
|
| 1572 |
+
"step": 221
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.32220609579100146,
|
| 1576 |
+
"grad_norm": 0.3737308084964752,
|
| 1577 |
+
"learning_rate": 0.00015562913435003114,
|
| 1578 |
+
"loss": 0.5868,
|
| 1579 |
+
"step": 222
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 0.32365747460087085,
|
| 1583 |
+
"grad_norm": 0.41030675172805786,
|
| 1584 |
+
"learning_rate": 0.00015524405962938714,
|
| 1585 |
+
"loss": 0.933,
|
| 1586 |
+
"step": 223
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.3251088534107402,
|
| 1590 |
+
"grad_norm": 0.22345803678035736,
|
| 1591 |
+
"learning_rate": 0.0001548578022893202,
|
| 1592 |
+
"loss": 0.4072,
|
| 1593 |
+
"step": 224
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.3265602322206096,
|
| 1597 |
+
"grad_norm": 0.4588911235332489,
|
| 1598 |
+
"learning_rate": 0.00015447037059851,
|
| 1599 |
+
"loss": 1.3702,
|
| 1600 |
+
"step": 225
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.32801161103047893,
|
| 1604 |
+
"grad_norm": 0.40378692746162415,
|
| 1605 |
+
"learning_rate": 0.00015408177285077565,
|
| 1606 |
+
"loss": 0.7234,
|
| 1607 |
+
"step": 226
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.32946298984034833,
|
| 1611 |
+
"grad_norm": 0.36572182178497314,
|
| 1612 |
+
"learning_rate": 0.0001536920173648984,
|
| 1613 |
+
"loss": 0.6561,
|
| 1614 |
+
"step": 227
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.3309143686502177,
|
| 1618 |
+
"grad_norm": 0.24225015938282013,
|
| 1619 |
+
"learning_rate": 0.0001533011124844433,
|
| 1620 |
+
"loss": 0.5723,
|
| 1621 |
+
"step": 228
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 0.33236574746008707,
|
| 1625 |
+
"grad_norm": 0.4880613386631012,
|
| 1626 |
+
"learning_rate": 0.00015290906657758083,
|
| 1627 |
+
"loss": 0.8659,
|
| 1628 |
+
"step": 229
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.33381712626995647,
|
| 1632 |
+
"grad_norm": 0.4412708878517151,
|
| 1633 |
+
"learning_rate": 0.00015251588803690753,
|
| 1634 |
+
"loss": 0.8392,
|
| 1635 |
+
"step": 230
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 0.3352685050798258,
|
| 1639 |
+
"grad_norm": 0.2982364594936371,
|
| 1640 |
+
"learning_rate": 0.00015212158527926634,
|
| 1641 |
+
"loss": 0.8567,
|
| 1642 |
+
"step": 231
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 0.3367198838896952,
|
| 1646 |
+
"grad_norm": 0.3174440860748291,
|
| 1647 |
+
"learning_rate": 0.0001517261667455667,
|
| 1648 |
+
"loss": 0.6775,
|
| 1649 |
+
"step": 232
|
| 1650 |
+
},
|
| 1651 |
+
{
|
| 1652 |
+
"epoch": 0.3381712626995646,
|
| 1653 |
+
"grad_norm": 0.23971304297447205,
|
| 1654 |
+
"learning_rate": 0.00015132964090060357,
|
| 1655 |
+
"loss": 0.2637,
|
| 1656 |
+
"step": 233
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 0.33962264150943394,
|
| 1660 |
+
"grad_norm": 0.3869999647140503,
|
| 1661 |
+
"learning_rate": 0.00015093201623287631,
|
| 1662 |
+
"loss": 0.7993,
|
| 1663 |
+
"step": 234
|
| 1664 |
+
},
|
| 1665 |
+
{
|
| 1666 |
+
"epoch": 0.34107402031930334,
|
| 1667 |
+
"grad_norm": 0.4646807014942169,
|
| 1668 |
+
"learning_rate": 0.000150533301254407,
|
| 1669 |
+
"loss": 1.0936,
|
| 1670 |
+
"step": 235
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.34252539912917274,
|
| 1674 |
+
"grad_norm": 0.39114901423454285,
|
| 1675 |
+
"learning_rate": 0.0001501335045005582,
|
| 1676 |
+
"loss": 1.0431,
|
| 1677 |
+
"step": 236
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 0.3439767779390421,
|
| 1681 |
+
"grad_norm": 0.32341787219047546,
|
| 1682 |
+
"learning_rate": 0.00014973263452985024,
|
| 1683 |
+
"loss": 0.5171,
|
| 1684 |
+
"step": 237
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 0.3454281567489115,
|
| 1688 |
+
"grad_norm": 0.3025743365287781,
|
| 1689 |
+
"learning_rate": 0.00014933069992377793,
|
| 1690 |
+
"loss": 0.59,
|
| 1691 |
+
"step": 238
|
| 1692 |
+
},
|
| 1693 |
+
{
|
| 1694 |
+
"epoch": 0.3468795355587808,
|
| 1695 |
+
"grad_norm": 0.49004557728767395,
|
| 1696 |
+
"learning_rate": 0.00014892770928662697,
|
| 1697 |
+
"loss": 1.6646,
|
| 1698 |
+
"step": 239
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 0.3483309143686502,
|
| 1702 |
+
"grad_norm": 0.5597532391548157,
|
| 1703 |
+
"learning_rate": 0.00014852367124528975,
|
| 1704 |
+
"loss": 1.3429,
|
| 1705 |
+
"step": 240
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.3497822931785196,
|
| 1709 |
+
"grad_norm": 0.4048459827899933,
|
| 1710 |
+
"learning_rate": 0.00014811859444908052,
|
| 1711 |
+
"loss": 1.4101,
|
| 1712 |
+
"step": 241
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.35123367198838895,
|
| 1716 |
+
"grad_norm": 0.5641282796859741,
|
| 1717 |
+
"learning_rate": 0.00014771248756955042,
|
| 1718 |
+
"loss": 1.6691,
|
| 1719 |
+
"step": 242
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 0.35268505079825835,
|
| 1723 |
+
"grad_norm": 0.4494577646255493,
|
| 1724 |
+
"learning_rate": 0.00014730535930030172,
|
| 1725 |
+
"loss": 1.211,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 0.35413642960812775,
|
| 1730 |
+
"grad_norm": 0.3181702792644501,
|
| 1731 |
+
"learning_rate": 0.00014689721835680182,
|
| 1732 |
+
"loss": 0.4831,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 0.3555878084179971,
|
| 1737 |
+
"grad_norm": 0.3166160583496094,
|
| 1738 |
+
"learning_rate": 0.00014648807347619663,
|
| 1739 |
+
"loss": 1.59,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.3570391872278665,
|
| 1744 |
+
"grad_norm": 0.3532291054725647,
|
| 1745 |
+
"learning_rate": 0.00014607793341712334,
|
| 1746 |
+
"loss": 0.3531,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 0.3584905660377358,
|
| 1751 |
+
"grad_norm": 0.3703453540802002,
|
| 1752 |
+
"learning_rate": 0.00014566680695952332,
|
| 1753 |
+
"loss": 0.7499,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.3599419448476052,
|
| 1758 |
+
"grad_norm": 0.29574841260910034,
|
| 1759 |
+
"learning_rate": 0.00014525470290445392,
|
| 1760 |
+
"loss": 0.7956,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 0.3613933236574746,
|
| 1765 |
+
"grad_norm": 0.41456523537635803,
|
| 1766 |
+
"learning_rate": 0.0001448416300738999,
|
| 1767 |
+
"loss": 1.2597,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 0.36284470246734396,
|
| 1772 |
+
"grad_norm": 0.3268081545829773,
|
| 1773 |
+
"learning_rate": 0.000144427597310585,
|
| 1774 |
+
"loss": 0.7744,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 0.36429608127721336,
|
| 1779 |
+
"grad_norm": 0.4419240951538086,
|
| 1780 |
+
"learning_rate": 0.00014401261347778233,
|
| 1781 |
+
"loss": 1.2126,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 0.36574746008708275,
|
| 1786 |
+
"grad_norm": 0.4968956410884857,
|
| 1787 |
+
"learning_rate": 0.0001435966874591247,
|
| 1788 |
+
"loss": 0.8786,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 0.3671988388969521,
|
| 1793 |
+
"grad_norm": 0.4824519455432892,
|
| 1794 |
+
"learning_rate": 0.0001431798281584144,
|
| 1795 |
+
"loss": 3.0825,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.3686502177068215,
|
| 1800 |
+
"grad_norm": 0.30820131301879883,
|
| 1801 |
+
"learning_rate": 0.0001427620444994328,
|
| 1802 |
+
"loss": 0.6051,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 0.37010159651669083,
|
| 1807 |
+
"grad_norm": 0.42164450883865356,
|
| 1808 |
+
"learning_rate": 0.00014234334542574906,
|
| 1809 |
+
"loss": 0.5161,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.37155297532656023,
|
| 1814 |
+
"grad_norm": 0.32636889815330505,
|
| 1815 |
+
"learning_rate": 0.00014192373990052877,
|
| 1816 |
+
"loss": 0.855,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.37300435413642963,
|
| 1821 |
+
"grad_norm": 0.332589715719223,
|
| 1822 |
+
"learning_rate": 0.0001415032369063422,
|
| 1823 |
+
"loss": 0.402,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.37445573294629897,
|
| 1828 |
+
"grad_norm": 0.37509143352508545,
|
| 1829 |
+
"learning_rate": 0.00014108184544497178,
|
| 1830 |
+
"loss": 1.1532,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.37590711175616837,
|
| 1835 |
+
"grad_norm": 0.2669030725955963,
|
| 1836 |
+
"learning_rate": 0.00014065957453721962,
|
| 1837 |
+
"loss": 0.5493,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.37735849056603776,
|
| 1842 |
+
"grad_norm": 0.387205570936203,
|
| 1843 |
+
"learning_rate": 0.00014023643322271426,
|
| 1844 |
+
"loss": 1.3697,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.3788098693759071,
|
| 1849 |
+
"grad_norm": 0.3195268213748932,
|
| 1850 |
+
"learning_rate": 0.00013981243055971712,
|
| 1851 |
+
"loss": 0.731,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 0.3802612481857765,
|
| 1856 |
+
"grad_norm": 0.3117748498916626,
|
| 1857 |
+
"learning_rate": 0.00013938757562492873,
|
| 1858 |
+
"loss": 0.4412,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 0.38171262699564584,
|
| 1863 |
+
"grad_norm": 0.43555042147636414,
|
| 1864 |
+
"learning_rate": 0.00013896187751329437,
|
| 1865 |
+
"loss": 0.9857,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.38316400580551524,
|
| 1870 |
+
"grad_norm": 0.3363146483898163,
|
| 1871 |
+
"learning_rate": 0.0001385353453378093,
|
| 1872 |
+
"loss": 0.8749,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.38461538461538464,
|
| 1877 |
+
"grad_norm": 0.3038422465324402,
|
| 1878 |
+
"learning_rate": 0.00013810798822932378,
|
| 1879 |
+
"loss": 0.5955,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.386066763425254,
|
| 1884 |
+
"grad_norm": 0.47416090965270996,
|
| 1885 |
+
"learning_rate": 0.00013767981533634754,
|
| 1886 |
+
"loss": 3.7495,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 0.3875181422351234,
|
| 1891 |
+
"grad_norm": 0.237370103597641,
|
| 1892 |
+
"learning_rate": 0.00013725083582485397,
|
| 1893 |
+
"loss": 0.3979,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 0.3889695210449927,
|
| 1898 |
+
"grad_norm": 0.1970636397600174,
|
| 1899 |
+
"learning_rate": 0.0001368210588780838,
|
| 1900 |
+
"loss": 0.297,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.3904208998548621,
|
| 1905 |
+
"grad_norm": 0.5174916982650757,
|
| 1906 |
+
"learning_rate": 0.00013639049369634876,
|
| 1907 |
+
"loss": 1.4907,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 0.3918722786647315,
|
| 1912 |
+
"grad_norm": 0.2811208665370941,
|
| 1913 |
+
"learning_rate": 0.00013595914949683432,
|
| 1914 |
+
"loss": 0.5772,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.39332365747460085,
|
| 1919 |
+
"grad_norm": 0.30495163798332214,
|
| 1920 |
+
"learning_rate": 0.00013552703551340258,
|
| 1921 |
+
"loss": 0.3624,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.39477503628447025,
|
| 1926 |
+
"grad_norm": 0.4540651738643646,
|
| 1927 |
+
"learning_rate": 0.00013509416099639457,
|
| 1928 |
+
"loss": 1.6049,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.39622641509433965,
|
| 1933 |
+
"grad_norm": 0.4534936547279358,
|
| 1934 |
+
"learning_rate": 0.00013466053521243214,
|
| 1935 |
+
"loss": 1.3386,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.397677793904209,
|
| 1940 |
+
"grad_norm": 0.23787882924079895,
|
| 1941 |
+
"learning_rate": 0.00013422616744421966,
|
| 1942 |
+
"loss": 0.5185,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.3991291727140784,
|
| 1947 |
+
"grad_norm": 0.35515129566192627,
|
| 1948 |
+
"learning_rate": 0.00013379106699034537,
|
| 1949 |
+
"loss": 1.4753,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.4005805515239477,
|
| 1954 |
+
"grad_norm": 0.38288307189941406,
|
| 1955 |
+
"learning_rate": 0.00013335524316508208,
|
| 1956 |
+
"loss": 1.4326,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 0.4020319303338171,
|
| 1961 |
+
"grad_norm": 0.27358075976371765,
|
| 1962 |
+
"learning_rate": 0.0001329187052981881,
|
| 1963 |
+
"loss": 0.683,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.4034833091436865,
|
| 1968 |
+
"grad_norm": 0.5965114831924438,
|
| 1969 |
+
"learning_rate": 0.00013248146273470726,
|
| 1970 |
+
"loss": 2.2199,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 0.40493468795355586,
|
| 1975 |
+
"grad_norm": 0.27985599637031555,
|
| 1976 |
+
"learning_rate": 0.00013204352483476897,
|
| 1977 |
+
"loss": 0.6224,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 0.40638606676342526,
|
| 1982 |
+
"grad_norm": 0.31146517395973206,
|
| 1983 |
+
"learning_rate": 0.0001316049009733879,
|
| 1984 |
+
"loss": 0.9722,
|
| 1985 |
+
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 0.40783744557329465,
|
| 1989 |
+
"grad_norm": 0.2670224905014038,
|
| 1990 |
+
"learning_rate": 0.0001311656005402631,
|
| 1991 |
+
"loss": 0.4706,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.409288824383164,
|
| 1996 |
+
"grad_norm": 0.30668875575065613,
|
| 1997 |
+
"learning_rate": 0.00013072563293957723,
|
| 1998 |
+
"loss": 0.8013,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.4107402031930334,
|
| 2003 |
+
"grad_norm": 1.6362825632095337,
|
| 2004 |
+
"learning_rate": 0.00013028500758979506,
|
| 2005 |
+
"loss": 1.7021,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.41219158200290273,
|
| 2010 |
+
"grad_norm": 0.3747856914997101,
|
| 2011 |
+
"learning_rate": 0.00012984373392346194,
|
| 2012 |
+
"loss": 0.7395,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 0.41364296081277213,
|
| 2017 |
+
"grad_norm": 0.29460155963897705,
|
| 2018 |
+
"learning_rate": 0.0001294018213870018,
|
| 2019 |
+
"loss": 0.4852,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.41509433962264153,
|
| 2024 |
+
"grad_norm": 0.3442453444004059,
|
| 2025 |
+
"learning_rate": 0.00012895927944051502,
|
| 2026 |
+
"loss": 1.0444,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.41654571843251087,
|
| 2031 |
+
"grad_norm": 0.4372914731502533,
|
| 2032 |
+
"learning_rate": 0.00012851611755757586,
|
| 2033 |
+
"loss": 0.9556,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.41799709724238027,
|
| 2038 |
+
"grad_norm": 0.36953088641166687,
|
| 2039 |
+
"learning_rate": 0.00012807234522502968,
|
| 2040 |
+
"loss": 0.6292,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.41944847605224966,
|
| 2045 |
+
"grad_norm": 0.2996106445789337,
|
| 2046 |
+
"learning_rate": 0.00012762797194278984,
|
| 2047 |
+
"loss": 0.4637,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.420899854862119,
|
| 2052 |
+
"grad_norm": 0.46969079971313477,
|
| 2053 |
+
"learning_rate": 0.0001271830072236343,
|
| 2054 |
+
"loss": 2.3145,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 0.4223512336719884,
|
| 2059 |
+
"grad_norm": 0.377268522977829,
|
| 2060 |
+
"learning_rate": 0.00012673746059300208,
|
| 2061 |
+
"loss": 1.1514,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.42380261248185774,
|
| 2066 |
+
"grad_norm": 0.37440070509910583,
|
| 2067 |
+
"learning_rate": 0.00012629134158878918,
|
| 2068 |
+
"loss": 1.2862,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 0.42525399129172714,
|
| 2073 |
+
"grad_norm": 0.47031837701797485,
|
| 2074 |
+
"learning_rate": 0.0001258446597611447,
|
| 2075 |
+
"loss": 1.3154,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 0.42670537010159654,
|
| 2080 |
+
"grad_norm": 0.28943926095962524,
|
| 2081 |
+
"learning_rate": 0.00012539742467226601,
|
| 2082 |
+
"loss": 0.7372,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 0.4281567489114659,
|
| 2087 |
+
"grad_norm": 0.3967966139316559,
|
| 2088 |
+
"learning_rate": 0.00012494964589619423,
|
| 2089 |
+
"loss": 1.7696,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.4296081277213353,
|
| 2094 |
+
"grad_norm": 0.3762686252593994,
|
| 2095 |
+
"learning_rate": 0.00012450133301860952,
|
| 2096 |
+
"loss": 1.2166,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 0.4310595065312046,
|
| 2101 |
+
"grad_norm": 0.3782839775085449,
|
| 2102 |
+
"learning_rate": 0.00012405249563662537,
|
| 2103 |
+
"loss": 0.9249,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 0.432510885341074,
|
| 2108 |
+
"grad_norm": 0.551977276802063,
|
| 2109 |
+
"learning_rate": 0.0001236031433585836,
|
| 2110 |
+
"loss": 0.7932,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.4339622641509434,
|
| 2115 |
+
"grad_norm": 0.4811376631259918,
|
| 2116 |
+
"learning_rate": 0.00012315328580384842,
|
| 2117 |
+
"loss": 1.4454,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 0.43541364296081275,
|
| 2122 |
+
"grad_norm": 0.6780052185058594,
|
| 2123 |
+
"learning_rate": 0.00012270293260260067,
|
| 2124 |
+
"loss": 1.8719,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 0.43686502177068215,
|
| 2129 |
+
"grad_norm": 0.5372028350830078,
|
| 2130 |
+
"learning_rate": 0.00012225209339563145,
|
| 2131 |
+
"loss": 1.13,
|
| 2132 |
+
"step": 301
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 0.43831640058055155,
|
| 2136 |
+
"grad_norm": 0.6659995317459106,
|
| 2137 |
+
"learning_rate": 0.00012180077783413601,
|
| 2138 |
+
"loss": 1.2524,
|
| 2139 |
+
"step": 302
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 0.4397677793904209,
|
| 2143 |
+
"grad_norm": 0.30327922105789185,
|
| 2144 |
+
"learning_rate": 0.00012134899557950698,
|
| 2145 |
+
"loss": 0.4541,
|
| 2146 |
+
"step": 303
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 0.4412191582002903,
|
| 2150 |
+
"grad_norm": 0.28171366453170776,
|
| 2151 |
+
"learning_rate": 0.00012089675630312754,
|
| 2152 |
+
"loss": 0.5961,
|
| 2153 |
+
"step": 304
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 0.4426705370101596,
|
| 2157 |
+
"grad_norm": 0.49443453550338745,
|
| 2158 |
+
"learning_rate": 0.00012044406968616432,
|
| 2159 |
+
"loss": 1.3738,
|
| 2160 |
+
"step": 305
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 0.444121915820029,
|
| 2164 |
+
"grad_norm": 0.27701106667518616,
|
| 2165 |
+
"learning_rate": 0.00011999094541936047,
|
| 2166 |
+
"loss": 0.8035,
|
| 2167 |
+
"step": 306
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 0.4455732946298984,
|
| 2171 |
+
"grad_norm": 0.44474315643310547,
|
| 2172 |
+
"learning_rate": 0.00011953739320282778,
|
| 2173 |
+
"loss": 1.9202,
|
| 2174 |
+
"step": 307
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 0.44702467343976776,
|
| 2178 |
+
"grad_norm": 0.36672115325927734,
|
| 2179 |
+
"learning_rate": 0.00011908342274583936,
|
| 2180 |
+
"loss": 1.3322,
|
| 2181 |
+
"step": 308
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 0.44847605224963716,
|
| 2185 |
+
"grad_norm": 0.47056931257247925,
|
| 2186 |
+
"learning_rate": 0.00011862904376662167,
|
| 2187 |
+
"loss": 0.6009,
|
| 2188 |
+
"step": 309
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 0.44992743105950656,
|
| 2192 |
+
"grad_norm": 0.33981776237487793,
|
| 2193 |
+
"learning_rate": 0.00011817426599214636,
|
| 2194 |
+
"loss": 0.8832,
|
| 2195 |
+
"step": 310
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 0.4513788098693759,
|
| 2199 |
+
"grad_norm": 0.5165224075317383,
|
| 2200 |
+
"learning_rate": 0.0001177190991579223,
|
| 2201 |
+
"loss": 1.2614,
|
| 2202 |
+
"step": 311
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 0.4528301886792453,
|
| 2206 |
+
"grad_norm": 0.36011219024658203,
|
| 2207 |
+
"learning_rate": 0.00011726355300778693,
|
| 2208 |
+
"loss": 0.5128,
|
| 2209 |
+
"step": 312
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 0.45428156748911463,
|
| 2213 |
+
"grad_norm": 0.4018263518810272,
|
| 2214 |
+
"learning_rate": 0.00011680763729369783,
|
| 2215 |
+
"loss": 0.876,
|
| 2216 |
+
"step": 313
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 0.45573294629898403,
|
| 2220 |
+
"grad_norm": 0.36778366565704346,
|
| 2221 |
+
"learning_rate": 0.00011635136177552391,
|
| 2222 |
+
"loss": 0.6217,
|
| 2223 |
+
"step": 314
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 0.45718432510885343,
|
| 2227 |
+
"grad_norm": 0.3395918905735016,
|
| 2228 |
+
"learning_rate": 0.00011589473622083642,
|
| 2229 |
+
"loss": 0.7758,
|
| 2230 |
+
"step": 315
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 0.45863570391872277,
|
| 2234 |
+
"grad_norm": 0.30903902649879456,
|
| 2235 |
+
"learning_rate": 0.00011543777040469994,
|
| 2236 |
+
"loss": 1.1969,
|
| 2237 |
+
"step": 316
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 0.46008708272859217,
|
| 2241 |
+
"grad_norm": 0.36154720187187195,
|
| 2242 |
+
"learning_rate": 0.00011498047410946306,
|
| 2243 |
+
"loss": 0.7883,
|
| 2244 |
+
"step": 317
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 0.46153846153846156,
|
| 2248 |
+
"grad_norm": 0.3478432595729828,
|
| 2249 |
+
"learning_rate": 0.00011452285712454904,
|
| 2250 |
+
"loss": 1.1028,
|
| 2251 |
+
"step": 318
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 0.4629898403483309,
|
| 2255 |
+
"grad_norm": 0.8324244618415833,
|
| 2256 |
+
"learning_rate": 0.00011406492924624614,
|
| 2257 |
+
"loss": 2.98,
|
| 2258 |
+
"step": 319
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 0.4644412191582003,
|
| 2262 |
+
"grad_norm": 0.36261022090911865,
|
| 2263 |
+
"learning_rate": 0.00011360670027749807,
|
| 2264 |
+
"loss": 0.8629,
|
| 2265 |
+
"step": 320
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 0.46589259796806964,
|
| 2269 |
+
"grad_norm": 0.42329132556915283,
|
| 2270 |
+
"learning_rate": 0.00011314818002769389,
|
| 2271 |
+
"loss": 1.8163,
|
| 2272 |
+
"step": 321
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 0.46734397677793904,
|
| 2276 |
+
"grad_norm": 0.5022541284561157,
|
| 2277 |
+
"learning_rate": 0.0001126893783124583,
|
| 2278 |
+
"loss": 1.5672,
|
| 2279 |
+
"step": 322
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 0.46879535558780844,
|
| 2283 |
+
"grad_norm": 0.3574770987033844,
|
| 2284 |
+
"learning_rate": 0.00011223030495344127,
|
| 2285 |
+
"loss": 0.3025,
|
| 2286 |
+
"step": 323
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 0.4702467343976778,
|
| 2290 |
+
"grad_norm": 0.4947843551635742,
|
| 2291 |
+
"learning_rate": 0.00011177096977810803,
|
| 2292 |
+
"loss": 0.8242,
|
| 2293 |
+
"step": 324
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 0.4716981132075472,
|
| 2297 |
+
"grad_norm": 0.4501658082008362,
|
| 2298 |
+
"learning_rate": 0.00011131138261952845,
|
| 2299 |
+
"loss": 0.8425,
|
| 2300 |
+
"step": 325
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 0.4731494920174166,
|
| 2304 |
+
"grad_norm": 0.21305899322032928,
|
| 2305 |
+
"learning_rate": 0.00011085155331616663,
|
| 2306 |
+
"loss": 0.245,
|
| 2307 |
+
"step": 326
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 0.4746008708272859,
|
| 2311 |
+
"grad_norm": 0.38023510575294495,
|
| 2312 |
+
"learning_rate": 0.00011039149171167045,
|
| 2313 |
+
"loss": 1.0404,
|
| 2314 |
+
"step": 327
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 0.4760522496371553,
|
| 2318 |
+
"grad_norm": 0.4849722385406494,
|
| 2319 |
+
"learning_rate": 0.00010993120765466056,
|
| 2320 |
+
"loss": 1.0257,
|
| 2321 |
+
"step": 328
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 0.47750362844702465,
|
| 2325 |
+
"grad_norm": 0.4337415099143982,
|
| 2326 |
+
"learning_rate": 0.00010947071099851971,
|
| 2327 |
+
"loss": 0.787,
|
| 2328 |
+
"step": 329
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 0.47895500725689405,
|
| 2332 |
+
"grad_norm": 0.3395763635635376,
|
| 2333 |
+
"learning_rate": 0.00010901001160118189,
|
| 2334 |
+
"loss": 0.9218,
|
| 2335 |
+
"step": 330
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 0.48040638606676345,
|
| 2339 |
+
"grad_norm": 0.5438852906227112,
|
| 2340 |
+
"learning_rate": 0.00010854911932492114,
|
| 2341 |
+
"loss": 0.7867,
|
| 2342 |
+
"step": 331
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 0.4818577648766328,
|
| 2346 |
+
"grad_norm": 0.299010306596756,
|
| 2347 |
+
"learning_rate": 0.00010808804403614043,
|
| 2348 |
+
"loss": 0.7583,
|
| 2349 |
+
"step": 332
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 0.4833091436865022,
|
| 2353 |
+
"grad_norm": 0.44489172101020813,
|
| 2354 |
+
"learning_rate": 0.00010762679560516067,
|
| 2355 |
+
"loss": 1.9207,
|
| 2356 |
+
"step": 333
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 0.4847605224963715,
|
| 2360 |
+
"grad_norm": 0.3161913752555847,
|
| 2361 |
+
"learning_rate": 0.00010716538390600908,
|
| 2362 |
+
"loss": 0.7175,
|
| 2363 |
+
"step": 334
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 0.4862119013062409,
|
| 2367 |
+
"grad_norm": 0.38730329275131226,
|
| 2368 |
+
"learning_rate": 0.00010670381881620814,
|
| 2369 |
+
"loss": 1.2888,
|
| 2370 |
+
"step": 335
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 0.4876632801161103,
|
| 2374 |
+
"grad_norm": 0.3712642788887024,
|
| 2375 |
+
"learning_rate": 0.00010624211021656392,
|
| 2376 |
+
"loss": 0.9692,
|
| 2377 |
+
"step": 336
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 0.48911465892597966,
|
| 2381 |
+
"grad_norm": 0.3168681859970093,
|
| 2382 |
+
"learning_rate": 0.00010578026799095464,
|
| 2383 |
+
"loss": 1.1365,
|
| 2384 |
+
"step": 337
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 0.49056603773584906,
|
| 2388 |
+
"grad_norm": 0.36398571729660034,
|
| 2389 |
+
"learning_rate": 0.00010531830202611904,
|
| 2390 |
+
"loss": 0.9235,
|
| 2391 |
+
"step": 338
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 0.49201741654571846,
|
| 2395 |
+
"grad_norm": 0.30887413024902344,
|
| 2396 |
+
"learning_rate": 0.00010485622221144484,
|
| 2397 |
+
"loss": 0.4427,
|
| 2398 |
+
"step": 339
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 0.4934687953555878,
|
| 2402 |
+
"grad_norm": 0.26171329617500305,
|
| 2403 |
+
"learning_rate": 0.0001043940384387569,
|
| 2404 |
+
"loss": 0.6965,
|
| 2405 |
+
"step": 340
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 0.4949201741654572,
|
| 2409 |
+
"grad_norm": 0.44587722420692444,
|
| 2410 |
+
"learning_rate": 0.00010393176060210557,
|
| 2411 |
+
"loss": 1.8459,
|
| 2412 |
+
"step": 341
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 0.49637155297532654,
|
| 2416 |
+
"grad_norm": 0.341634064912796,
|
| 2417 |
+
"learning_rate": 0.0001034693985975548,
|
| 2418 |
+
"loss": 0.7757,
|
| 2419 |
+
"step": 342
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 0.49782293178519593,
|
| 2423 |
+
"grad_norm": 0.378426194190979,
|
| 2424 |
+
"learning_rate": 0.0001030069623229704,
|
| 2425 |
+
"loss": 1.1976,
|
| 2426 |
+
"step": 343
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 0.49927431059506533,
|
| 2430 |
+
"grad_norm": 0.4587966799736023,
|
| 2431 |
+
"learning_rate": 0.00010254446167780803,
|
| 2432 |
+
"loss": 0.7019,
|
| 2433 |
+
"step": 344
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 0.5007256894049347,
|
| 2437 |
+
"grad_norm": 0.2761523127555847,
|
| 2438 |
+
"learning_rate": 0.00010208190656290137,
|
| 2439 |
+
"loss": 0.3916,
|
| 2440 |
+
"step": 345
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 0.502177068214804,
|
| 2444 |
+
"grad_norm": 0.42763715982437134,
|
| 2445 |
+
"learning_rate": 0.00010161930688025017,
|
| 2446 |
+
"loss": 0.8072,
|
| 2447 |
+
"step": 346
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 0.502177068214804,
|
| 2451 |
+
"eval_loss": 0.23982280492782593,
|
| 2452 |
+
"eval_runtime": 32.4007,
|
| 2453 |
+
"eval_samples_per_second": 8.981,
|
| 2454 |
+
"eval_steps_per_second": 4.506,
|
| 2455 |
+
"step": 346
|
| 2456 |
}
|
| 2457 |
],
|
| 2458 |
"logging_steps": 1,
|
|
|
|
| 2472 |
"attributes": {}
|
| 2473 |
}
|
| 2474 |
},
|
| 2475 |
+
"total_flos": 2.1878609950015488e+17,
|
| 2476 |
"train_batch_size": 2,
|
| 2477 |
"trial_name": null,
|
| 2478 |
"trial_params": null
|