Training in progress, step 436, 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 614968888
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3eabb358380da88f3e2eb2384a4f4f8f13aedb6194ee219a6d4e44932060672
|
| 3 |
size 614968888
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1230132424
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d1d80a8bc8236676e6232fb2db29ec0a42aa6330168f2aaf72934c21e75f0d2
|
| 3 |
size 1230132424
|
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:d3d44e1ca4f3d5f0b30777072074e21fca5cb6c135bd86fac71f54a60249896c
|
| 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 sha256:4afd94ddd3630727d7e4646bbec3048ec7f0e8b826d6c96ce73546e36a1353bf
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1680,6 +1680,1385 @@
|
|
| 1680 |
"learning_rate": 6.339194928502517e-05,
|
| 1681 |
"loss": 1.9865,
|
| 1682 |
"step": 239
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1683 |
}
|
| 1684 |
],
|
| 1685 |
"logging_steps": 1,
|
|
@@ -1694,12 +3073,12 @@
|
|
| 1694 |
"should_evaluate": false,
|
| 1695 |
"should_log": false,
|
| 1696 |
"should_save": true,
|
| 1697 |
-
"should_training_stop":
|
| 1698 |
},
|
| 1699 |
"attributes": {}
|
| 1700 |
}
|
| 1701 |
},
|
| 1702 |
-
"total_flos":
|
| 1703 |
"train_batch_size": 4,
|
| 1704 |
"trial_name": null,
|
| 1705 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 436,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1680 |
"learning_rate": 6.339194928502517e-05,
|
| 1681 |
"loss": 1.9865,
|
| 1682 |
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.5504587155963303,
|
| 1686 |
+
"grad_norm": 8.680039405822754,
|
| 1687 |
+
"learning_rate": 6.294095225512603e-05,
|
| 1688 |
+
"loss": 0.8176,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.5527522935779816,
|
| 1693 |
+
"grad_norm": 21.01581382751465,
|
| 1694 |
+
"learning_rate": 6.248882390836135e-05,
|
| 1695 |
+
"loss": 0.7879,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.555045871559633,
|
| 1700 |
+
"grad_norm": 10.070744514465332,
|
| 1701 |
+
"learning_rate": 6.203560377044866e-05,
|
| 1702 |
+
"loss": 0.8343,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.5573394495412844,
|
| 1707 |
+
"grad_norm": 3.3140993118286133,
|
| 1708 |
+
"learning_rate": 6.158133146255153e-05,
|
| 1709 |
+
"loss": 0.4982,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.5596330275229358,
|
| 1714 |
+
"grad_norm": 5.305646896362305,
|
| 1715 |
+
"learning_rate": 6.112604669781572e-05,
|
| 1716 |
+
"loss": 0.6298,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.5619266055045872,
|
| 1721 |
+
"grad_norm": 3.5618624687194824,
|
| 1722 |
+
"learning_rate": 6.0669789277897507e-05,
|
| 1723 |
+
"loss": 0.4889,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.5642201834862385,
|
| 1728 |
+
"grad_norm": 4.08172082901001,
|
| 1729 |
+
"learning_rate": 6.021259908948402e-05,
|
| 1730 |
+
"loss": 0.5831,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.5665137614678899,
|
| 1735 |
+
"grad_norm": 3.6568925380706787,
|
| 1736 |
+
"learning_rate": 5.9754516100806423e-05,
|
| 1737 |
+
"loss": 0.6037,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.5688073394495413,
|
| 1742 |
+
"grad_norm": 3.1750009059906006,
|
| 1743 |
+
"learning_rate": 5.9295580358145744e-05,
|
| 1744 |
+
"loss": 0.6221,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5711009174311926,
|
| 1749 |
+
"grad_norm": 3.9468281269073486,
|
| 1750 |
+
"learning_rate": 5.8835831982332015e-05,
|
| 1751 |
+
"loss": 0.6358,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.573394495412844,
|
| 1756 |
+
"grad_norm": 5.328709602355957,
|
| 1757 |
+
"learning_rate": 5.837531116523682e-05,
|
| 1758 |
+
"loss": 0.7811,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5756880733944955,
|
| 1763 |
+
"grad_norm": 1.4755908250808716,
|
| 1764 |
+
"learning_rate": 5.791405816625975e-05,
|
| 1765 |
+
"loss": 0.5937,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.5779816513761468,
|
| 1770 |
+
"grad_norm": 1.5495307445526123,
|
| 1771 |
+
"learning_rate": 5.745211330880872e-05,
|
| 1772 |
+
"loss": 0.589,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5802752293577982,
|
| 1777 |
+
"grad_norm": 1.6908055543899536,
|
| 1778 |
+
"learning_rate": 5.698951697677498e-05,
|
| 1779 |
+
"loss": 0.6837,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5825688073394495,
|
| 1784 |
+
"grad_norm": 1.8224513530731201,
|
| 1785 |
+
"learning_rate": 5.6526309611002594e-05,
|
| 1786 |
+
"loss": 0.9065,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.5848623853211009,
|
| 1791 |
+
"grad_norm": 1.8069437742233276,
|
| 1792 |
+
"learning_rate": 5.6062531705753075e-05,
|
| 1793 |
+
"loss": 0.7501,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.5871559633027523,
|
| 1798 |
+
"grad_norm": 2.2407689094543457,
|
| 1799 |
+
"learning_rate": 5.559822380516539e-05,
|
| 1800 |
+
"loss": 0.9369,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.5894495412844036,
|
| 1805 |
+
"grad_norm": 2.168276071548462,
|
| 1806 |
+
"learning_rate": 5.5133426499711425e-05,
|
| 1807 |
+
"loss": 1.0019,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.591743119266055,
|
| 1812 |
+
"grad_norm": 3.047189474105835,
|
| 1813 |
+
"learning_rate": 5.466818042264753e-05,
|
| 1814 |
+
"loss": 1.3919,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.5940366972477065,
|
| 1819 |
+
"grad_norm": 2.7798616886138916,
|
| 1820 |
+
"learning_rate": 5.420252624646238e-05,
|
| 1821 |
+
"loss": 1.1665,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.5963302752293578,
|
| 1826 |
+
"grad_norm": 3.3886473178863525,
|
| 1827 |
+
"learning_rate": 5.373650467932122e-05,
|
| 1828 |
+
"loss": 1.6027,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.5986238532110092,
|
| 1833 |
+
"grad_norm": 3.44380784034729,
|
| 1834 |
+
"learning_rate": 5.327015646150716e-05,
|
| 1835 |
+
"loss": 1.3152,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6009174311926605,
|
| 1840 |
+
"grad_norm": 3.924863338470459,
|
| 1841 |
+
"learning_rate": 5.2803522361859594e-05,
|
| 1842 |
+
"loss": 1.2854,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6032110091743119,
|
| 1847 |
+
"grad_norm": 3.6491658687591553,
|
| 1848 |
+
"learning_rate": 5.233664317421012e-05,
|
| 1849 |
+
"loss": 1.0868,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6055045871559633,
|
| 1854 |
+
"grad_norm": 4.161890983581543,
|
| 1855 |
+
"learning_rate": 5.18695597138163e-05,
|
| 1856 |
+
"loss": 1.4397,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6077981651376146,
|
| 1861 |
+
"grad_norm": 5.066677570343018,
|
| 1862 |
+
"learning_rate": 5.140231281379345e-05,
|
| 1863 |
+
"loss": 1.5409,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6100917431192661,
|
| 1868 |
+
"grad_norm": 5.091641426086426,
|
| 1869 |
+
"learning_rate": 5.0934943321545115e-05,
|
| 1870 |
+
"loss": 1.6991,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6123853211009175,
|
| 1875 |
+
"grad_norm": 6.486395835876465,
|
| 1876 |
+
"learning_rate": 5.046749209519197e-05,
|
| 1877 |
+
"loss": 1.7857,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6146788990825688,
|
| 1882 |
+
"grad_norm": 9.023844718933105,
|
| 1883 |
+
"learning_rate": 5e-05,
|
| 1884 |
+
"loss": 2.5128,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6169724770642202,
|
| 1889 |
+
"grad_norm": 6.766034126281738,
|
| 1890 |
+
"learning_rate": 4.953250790480805e-05,
|
| 1891 |
+
"loss": 1.9218,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6192660550458715,
|
| 1896 |
+
"grad_norm": 8.038894653320312,
|
| 1897 |
+
"learning_rate": 4.9065056678454904e-05,
|
| 1898 |
+
"loss": 2.2512,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6215596330275229,
|
| 1903 |
+
"grad_norm": 7.6417365074157715,
|
| 1904 |
+
"learning_rate": 4.859768718620656e-05,
|
| 1905 |
+
"loss": 2.4227,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6238532110091743,
|
| 1910 |
+
"grad_norm": 7.12877082824707,
|
| 1911 |
+
"learning_rate": 4.813044028618373e-05,
|
| 1912 |
+
"loss": 2.0528,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6261467889908257,
|
| 1917 |
+
"grad_norm": 7.904975891113281,
|
| 1918 |
+
"learning_rate": 4.76633568257899e-05,
|
| 1919 |
+
"loss": 2.2547,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.6284403669724771,
|
| 1924 |
+
"grad_norm": 6.577536582946777,
|
| 1925 |
+
"learning_rate": 4.7196477638140404e-05,
|
| 1926 |
+
"loss": 1.984,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.6307339449541285,
|
| 1931 |
+
"grad_norm": 8.355478286743164,
|
| 1932 |
+
"learning_rate": 4.6729843538492847e-05,
|
| 1933 |
+
"loss": 2.227,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.6330275229357798,
|
| 1938 |
+
"grad_norm": 8.487154006958008,
|
| 1939 |
+
"learning_rate": 4.626349532067879e-05,
|
| 1940 |
+
"loss": 2.1905,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.6353211009174312,
|
| 1945 |
+
"grad_norm": 8.718094825744629,
|
| 1946 |
+
"learning_rate": 4.579747375353763e-05,
|
| 1947 |
+
"loss": 2.0898,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.6376146788990825,
|
| 1952 |
+
"grad_norm": 9.59514045715332,
|
| 1953 |
+
"learning_rate": 4.5331819577352474e-05,
|
| 1954 |
+
"loss": 2.1952,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.6399082568807339,
|
| 1959 |
+
"grad_norm": 6.1734538078308105,
|
| 1960 |
+
"learning_rate": 4.486657350028859e-05,
|
| 1961 |
+
"loss": 1.4662,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.6422018348623854,
|
| 1966 |
+
"grad_norm": 8.729260444641113,
|
| 1967 |
+
"learning_rate": 4.4401776194834613e-05,
|
| 1968 |
+
"loss": 2.3723,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.6444954128440367,
|
| 1973 |
+
"grad_norm": 9.038458824157715,
|
| 1974 |
+
"learning_rate": 4.393746829424693e-05,
|
| 1975 |
+
"loss": 1.5177,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.6467889908256881,
|
| 1980 |
+
"grad_norm": 10.071760177612305,
|
| 1981 |
+
"learning_rate": 4.347369038899744e-05,
|
| 1982 |
+
"loss": 1.8683,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.6490825688073395,
|
| 1987 |
+
"grad_norm": 9.053478240966797,
|
| 1988 |
+
"learning_rate": 4.3010483023225045e-05,
|
| 1989 |
+
"loss": 1.7053,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.6513761467889908,
|
| 1994 |
+
"grad_norm": 8.493107795715332,
|
| 1995 |
+
"learning_rate": 4.254788669119127e-05,
|
| 1996 |
+
"loss": 0.8987,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.6536697247706422,
|
| 2001 |
+
"grad_norm": 9.247946739196777,
|
| 2002 |
+
"learning_rate": 4.208594183374026e-05,
|
| 2003 |
+
"loss": 0.7432,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.6559633027522935,
|
| 2008 |
+
"grad_norm": 7.0636115074157715,
|
| 2009 |
+
"learning_rate": 4.162468883476319e-05,
|
| 2010 |
+
"loss": 0.6019,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.658256880733945,
|
| 2015 |
+
"grad_norm": 11.127235412597656,
|
| 2016 |
+
"learning_rate": 4.1164168017668e-05,
|
| 2017 |
+
"loss": 2.198,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.6605504587155964,
|
| 2022 |
+
"grad_norm": 13.40218448638916,
|
| 2023 |
+
"learning_rate": 4.0704419641854274e-05,
|
| 2024 |
+
"loss": 2.4764,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.6628440366972477,
|
| 2029 |
+
"grad_norm": 6.9688029289245605,
|
| 2030 |
+
"learning_rate": 4.0245483899193595e-05,
|
| 2031 |
+
"loss": 1.2261,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6651376146788991,
|
| 2036 |
+
"grad_norm": 20.206615447998047,
|
| 2037 |
+
"learning_rate": 3.978740091051599e-05,
|
| 2038 |
+
"loss": 2.6922,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.6674311926605505,
|
| 2043 |
+
"grad_norm": 9.008017539978027,
|
| 2044 |
+
"learning_rate": 3.933021072210251e-05,
|
| 2045 |
+
"loss": 1.2186,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.6697247706422018,
|
| 2050 |
+
"grad_norm": 53.592010498046875,
|
| 2051 |
+
"learning_rate": 3.887395330218429e-05,
|
| 2052 |
+
"loss": 0.7876,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.6720183486238532,
|
| 2057 |
+
"grad_norm": 5.339064121246338,
|
| 2058 |
+
"learning_rate": 3.8418668537448495e-05,
|
| 2059 |
+
"loss": 0.6078,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.6743119266055045,
|
| 2064 |
+
"grad_norm": 3.749335765838623,
|
| 2065 |
+
"learning_rate": 3.7964396229551364e-05,
|
| 2066 |
+
"loss": 0.4519,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.676605504587156,
|
| 2071 |
+
"grad_norm": 4.613084316253662,
|
| 2072 |
+
"learning_rate": 3.7511176091638653e-05,
|
| 2073 |
+
"loss": 0.5232,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.6788990825688074,
|
| 2078 |
+
"grad_norm": 4.828958511352539,
|
| 2079 |
+
"learning_rate": 3.705904774487396e-05,
|
| 2080 |
+
"loss": 0.4491,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.6811926605504587,
|
| 2085 |
+
"grad_norm": 3.427786111831665,
|
| 2086 |
+
"learning_rate": 3.660805071497485e-05,
|
| 2087 |
+
"loss": 0.4936,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.6834862385321101,
|
| 2092 |
+
"grad_norm": 3.4336676597595215,
|
| 2093 |
+
"learning_rate": 3.6158224428757535e-05,
|
| 2094 |
+
"loss": 0.5795,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.6857798165137615,
|
| 2099 |
+
"grad_norm": 3.517413377761841,
|
| 2100 |
+
"learning_rate": 3.5709608210690125e-05,
|
| 2101 |
+
"loss": 0.5185,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.6880733944954128,
|
| 2106 |
+
"grad_norm": 4.021430969238281,
|
| 2107 |
+
"learning_rate": 3.5262241279454785e-05,
|
| 2108 |
+
"loss": 0.7537,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.6903669724770642,
|
| 2113 |
+
"grad_norm": 1.3459668159484863,
|
| 2114 |
+
"learning_rate": 3.4816162744519263e-05,
|
| 2115 |
+
"loss": 0.7576,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.6926605504587156,
|
| 2120 |
+
"grad_norm": 1.2463675737380981,
|
| 2121 |
+
"learning_rate": 3.4371411602717784e-05,
|
| 2122 |
+
"loss": 0.4637,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.694954128440367,
|
| 2127 |
+
"grad_norm": 1.3952363729476929,
|
| 2128 |
+
"learning_rate": 3.392802673484193e-05,
|
| 2129 |
+
"loss": 0.6165,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.6972477064220184,
|
| 2134 |
+
"grad_norm": 1.5832312107086182,
|
| 2135 |
+
"learning_rate": 3.3486046902241664e-05,
|
| 2136 |
+
"loss": 0.7459,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.6995412844036697,
|
| 2141 |
+
"grad_norm": 1.7343112230300903,
|
| 2142 |
+
"learning_rate": 3.3045510743436665e-05,
|
| 2143 |
+
"loss": 0.785,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7018348623853211,
|
| 2148 |
+
"grad_norm": 2.200110673904419,
|
| 2149 |
+
"learning_rate": 3.2606456770738636e-05,
|
| 2150 |
+
"loss": 1.1081,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7041284403669725,
|
| 2155 |
+
"grad_norm": 1.953981637954712,
|
| 2156 |
+
"learning_rate": 3.216892336688435e-05,
|
| 2157 |
+
"loss": 0.8974,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7064220183486238,
|
| 2162 |
+
"grad_norm": 2.165806293487549,
|
| 2163 |
+
"learning_rate": 3.173294878168025e-05,
|
| 2164 |
+
"loss": 1.0284,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7087155963302753,
|
| 2169 |
+
"grad_norm": 1.6866214275360107,
|
| 2170 |
+
"learning_rate": 3.129857112865859e-05,
|
| 2171 |
+
"loss": 0.6057,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7110091743119266,
|
| 2176 |
+
"grad_norm": 2.105428457260132,
|
| 2177 |
+
"learning_rate": 3.086582838174551e-05,
|
| 2178 |
+
"loss": 0.7096,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.713302752293578,
|
| 2183 |
+
"grad_norm": 3.2524518966674805,
|
| 2184 |
+
"learning_rate": 3.0434758371941258e-05,
|
| 2185 |
+
"loss": 1.5531,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.7155963302752294,
|
| 2190 |
+
"grad_norm": 2.998818874359131,
|
| 2191 |
+
"learning_rate": 3.000539878401296e-05,
|
| 2192 |
+
"loss": 1.1911,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.7178899082568807,
|
| 2197 |
+
"grad_norm": 3.267188549041748,
|
| 2198 |
+
"learning_rate": 2.9577787153200197e-05,
|
| 2199 |
+
"loss": 1.2269,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7201834862385321,
|
| 2204 |
+
"grad_norm": 3.965543270111084,
|
| 2205 |
+
"learning_rate": 2.9151960861933614e-05,
|
| 2206 |
+
"loss": 1.5541,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.7224770642201835,
|
| 2211 |
+
"grad_norm": 4.665194511413574,
|
| 2212 |
+
"learning_rate": 2.8727957136566823e-05,
|
| 2213 |
+
"loss": 2.0018,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.7247706422018348,
|
| 2218 |
+
"grad_norm": 4.473689079284668,
|
| 2219 |
+
"learning_rate": 2.8305813044122097e-05,
|
| 2220 |
+
"loss": 1.6075,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.7270642201834863,
|
| 2225 |
+
"grad_norm": 5.812178611755371,
|
| 2226 |
+
"learning_rate": 2.7885565489049946e-05,
|
| 2227 |
+
"loss": 2.3831,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.7293577981651376,
|
| 2232 |
+
"grad_norm": 6.922824382781982,
|
| 2233 |
+
"learning_rate": 2.746725121000273e-05,
|
| 2234 |
+
"loss": 2.4602,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.731651376146789,
|
| 2239 |
+
"grad_norm": 7.8420329093933105,
|
| 2240 |
+
"learning_rate": 2.705090677662311e-05,
|
| 2241 |
+
"loss": 2.3614,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.7339449541284404,
|
| 2246 |
+
"grad_norm": 7.710655689239502,
|
| 2247 |
+
"learning_rate": 2.66365685863469e-05,
|
| 2248 |
+
"loss": 2.6817,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.7362385321100917,
|
| 2253 |
+
"grad_norm": 6.691256999969482,
|
| 2254 |
+
"learning_rate": 2.6224272861221243e-05,
|
| 2255 |
+
"loss": 1.5446,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.7385321100917431,
|
| 2260 |
+
"grad_norm": 8.001251220703125,
|
| 2261 |
+
"learning_rate": 2.581405564473801e-05,
|
| 2262 |
+
"loss": 2.3128,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.7408256880733946,
|
| 2267 |
+
"grad_norm": 7.6729583740234375,
|
| 2268 |
+
"learning_rate": 2.5405952798682842e-05,
|
| 2269 |
+
"loss": 2.1427,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.7431192660550459,
|
| 2274 |
+
"grad_norm": 7.609097957611084,
|
| 2275 |
+
"learning_rate": 2.500000000000001e-05,
|
| 2276 |
+
"loss": 1.8877,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.7454128440366973,
|
| 2281 |
+
"grad_norm": 7.955555438995361,
|
| 2282 |
+
"learning_rate": 2.459623273767354e-05,
|
| 2283 |
+
"loss": 2.0801,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.7477064220183486,
|
| 2288 |
+
"grad_norm": 8.366448402404785,
|
| 2289 |
+
"learning_rate": 2.4194686309624663e-05,
|
| 2290 |
+
"loss": 2.079,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.75,
|
| 2295 |
+
"grad_norm": 7.975131511688232,
|
| 2296 |
+
"learning_rate": 2.3795395819626114e-05,
|
| 2297 |
+
"loss": 1.9748,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.7522935779816514,
|
| 2302 |
+
"grad_norm": 8.098983764648438,
|
| 2303 |
+
"learning_rate": 2.3398396174233178e-05,
|
| 2304 |
+
"loss": 2.0463,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.7545871559633027,
|
| 2309 |
+
"grad_norm": 8.904077529907227,
|
| 2310 |
+
"learning_rate": 2.300372207973219e-05,
|
| 2311 |
+
"loss": 2.1938,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.7568807339449541,
|
| 2316 |
+
"grad_norm": 11.033387184143066,
|
| 2317 |
+
"learning_rate": 2.261140803910644e-05,
|
| 2318 |
+
"loss": 1.4517,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7591743119266054,
|
| 2323 |
+
"grad_norm": 9.668691635131836,
|
| 2324 |
+
"learning_rate": 2.2221488349019903e-05,
|
| 2325 |
+
"loss": 1.9265,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.7614678899082569,
|
| 2330 |
+
"grad_norm": 6.776542663574219,
|
| 2331 |
+
"learning_rate": 2.1833997096818898e-05,
|
| 2332 |
+
"loss": 1.1323,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.7637614678899083,
|
| 2337 |
+
"grad_norm": 7.969955921173096,
|
| 2338 |
+
"learning_rate": 2.144896815755224e-05,
|
| 2339 |
+
"loss": 1.5859,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.7660550458715596,
|
| 2344 |
+
"grad_norm": 8.130461692810059,
|
| 2345 |
+
"learning_rate": 2.1066435191009715e-05,
|
| 2346 |
+
"loss": 1.4637,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.768348623853211,
|
| 2351 |
+
"grad_norm": 8.98421573638916,
|
| 2352 |
+
"learning_rate": 2.0686431638779564e-05,
|
| 2353 |
+
"loss": 1.7904,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.7706422018348624,
|
| 2358 |
+
"grad_norm": 11.332923889160156,
|
| 2359 |
+
"learning_rate": 2.0308990721324927e-05,
|
| 2360 |
+
"loss": 2.4356,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.7729357798165137,
|
| 2365 |
+
"grad_norm": 12.836410522460938,
|
| 2366 |
+
"learning_rate": 1.9934145435079702e-05,
|
| 2367 |
+
"loss": 2.4158,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.7752293577981652,
|
| 2372 |
+
"grad_norm": 8.298335075378418,
|
| 2373 |
+
"learning_rate": 1.9561928549563968e-05,
|
| 2374 |
+
"loss": 1.2192,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.7775229357798165,
|
| 2379 |
+
"grad_norm": 15.296181678771973,
|
| 2380 |
+
"learning_rate": 1.9192372604519127e-05,
|
| 2381 |
+
"loss": 1.7368,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.7798165137614679,
|
| 2386 |
+
"grad_norm": 29.77173614501953,
|
| 2387 |
+
"learning_rate": 1.8825509907063327e-05,
|
| 2388 |
+
"loss": 2.0105,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.7821100917431193,
|
| 2393 |
+
"grad_norm": 52.93706512451172,
|
| 2394 |
+
"learning_rate": 1.8461372528867093e-05,
|
| 2395 |
+
"loss": 0.9187,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.7844036697247706,
|
| 2400 |
+
"grad_norm": 9.151922225952148,
|
| 2401 |
+
"learning_rate": 1.8099992303349577e-05,
|
| 2402 |
+
"loss": 0.4775,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.786697247706422,
|
| 2407 |
+
"grad_norm": 4.057742595672607,
|
| 2408 |
+
"learning_rate": 1.774140082289563e-05,
|
| 2409 |
+
"loss": 0.4805,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.7889908256880734,
|
| 2414 |
+
"grad_norm": 5.886656284332275,
|
| 2415 |
+
"learning_rate": 1.738562943609396e-05,
|
| 2416 |
+
"loss": 0.5288,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.7912844036697247,
|
| 2421 |
+
"grad_norm": 3.6186325550079346,
|
| 2422 |
+
"learning_rate": 1.703270924499656e-05,
|
| 2423 |
+
"loss": 0.471,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.7935779816513762,
|
| 2428 |
+
"grad_norm": 3.7122743129730225,
|
| 2429 |
+
"learning_rate": 1.6682671102399805e-05,
|
| 2430 |
+
"loss": 0.4429,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.7958715596330275,
|
| 2435 |
+
"grad_norm": 3.1317362785339355,
|
| 2436 |
+
"learning_rate": 1.6335545609147142e-05,
|
| 2437 |
+
"loss": 0.5182,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.7981651376146789,
|
| 2442 |
+
"grad_norm": 3.575540065765381,
|
| 2443 |
+
"learning_rate": 1.599136311145402e-05,
|
| 2444 |
+
"loss": 0.6401,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.8004587155963303,
|
| 2449 |
+
"grad_norm": 3.384408712387085,
|
| 2450 |
+
"learning_rate": 1.5650153698254916e-05,
|
| 2451 |
+
"loss": 0.5007,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.8027522935779816,
|
| 2456 |
+
"grad_norm": 3.7142014503479004,
|
| 2457 |
+
"learning_rate": 1.531194719857292e-05,
|
| 2458 |
+
"loss": 0.5944,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.805045871559633,
|
| 2463 |
+
"grad_norm": 1.2152124643325806,
|
| 2464 |
+
"learning_rate": 1.4976773178912084e-05,
|
| 2465 |
+
"loss": 0.5656,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.8073394495412844,
|
| 2470 |
+
"grad_norm": 1.2784579992294312,
|
| 2471 |
+
"learning_rate": 1.4644660940672627e-05,
|
| 2472 |
+
"loss": 0.5921,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.8096330275229358,
|
| 2477 |
+
"grad_norm": 1.2699741125106812,
|
| 2478 |
+
"learning_rate": 1.4315639517589397e-05,
|
| 2479 |
+
"loss": 0.485,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.8119266055045872,
|
| 2484 |
+
"grad_norm": 1.3509347438812256,
|
| 2485 |
+
"learning_rate": 1.398973767319368e-05,
|
| 2486 |
+
"loss": 0.5775,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.8142201834862385,
|
| 2491 |
+
"grad_norm": 2.171257495880127,
|
| 2492 |
+
"learning_rate": 1.3666983898298657e-05,
|
| 2493 |
+
"loss": 1.2049,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.8165137614678899,
|
| 2498 |
+
"grad_norm": 1.8562023639678955,
|
| 2499 |
+
"learning_rate": 1.3347406408508695e-05,
|
| 2500 |
+
"loss": 0.8346,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.8188073394495413,
|
| 2505 |
+
"grad_norm": 1.9856826066970825,
|
| 2506 |
+
"learning_rate": 1.3031033141752702e-05,
|
| 2507 |
+
"loss": 0.8304,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8211009174311926,
|
| 2512 |
+
"grad_norm": 2.337763547897339,
|
| 2513 |
+
"learning_rate": 1.2717891755841722e-05,
|
| 2514 |
+
"loss": 1.2159,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.823394495412844,
|
| 2519 |
+
"grad_norm": 2.3222556114196777,
|
| 2520 |
+
"learning_rate": 1.2408009626051137e-05,
|
| 2521 |
+
"loss": 1.0698,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.8256880733944955,
|
| 2526 |
+
"grad_norm": 2.8859426975250244,
|
| 2527 |
+
"learning_rate": 1.2101413842727345e-05,
|
| 2528 |
+
"loss": 1.46,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.8279816513761468,
|
| 2533 |
+
"grad_norm": 2.7983813285827637,
|
| 2534 |
+
"learning_rate": 1.1798131208919627e-05,
|
| 2535 |
+
"loss": 1.2568,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.8302752293577982,
|
| 2540 |
+
"grad_norm": 3.1168298721313477,
|
| 2541 |
+
"learning_rate": 1.1498188238036861e-05,
|
| 2542 |
+
"loss": 1.3348,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.8325688073394495,
|
| 2547 |
+
"grad_norm": 3.040313482284546,
|
| 2548 |
+
"learning_rate": 1.1201611151529756e-05,
|
| 2549 |
+
"loss": 0.9272,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.8348623853211009,
|
| 2554 |
+
"grad_norm": 3.5628716945648193,
|
| 2555 |
+
"learning_rate": 1.090842587659851e-05,
|
| 2556 |
+
"loss": 1.1717,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8371559633027523,
|
| 2561 |
+
"grad_norm": 4.757216453552246,
|
| 2562 |
+
"learning_rate": 1.0618658043926233e-05,
|
| 2563 |
+
"loss": 1.7844,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.8394495412844036,
|
| 2568 |
+
"grad_norm": 5.807982444763184,
|
| 2569 |
+
"learning_rate": 1.0332332985438248e-05,
|
| 2570 |
+
"loss": 1.8459,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.841743119266055,
|
| 2575 |
+
"grad_norm": 6.931635856628418,
|
| 2576 |
+
"learning_rate": 1.004947573208756e-05,
|
| 2577 |
+
"loss": 2.4206,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.8440366972477065,
|
| 2582 |
+
"grad_norm": 6.323002338409424,
|
| 2583 |
+
"learning_rate": 9.770111011666583e-06,
|
| 2584 |
+
"loss": 1.7975,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.8463302752293578,
|
| 2589 |
+
"grad_norm": 6.554073333740234,
|
| 2590 |
+
"learning_rate": 9.494263246645474e-06,
|
| 2591 |
+
"loss": 2.1683,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.8486238532110092,
|
| 2596 |
+
"grad_norm": 7.123667240142822,
|
| 2597 |
+
"learning_rate": 9.221956552036992e-06,
|
| 2598 |
+
"loss": 2.4647,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.8509174311926605,
|
| 2603 |
+
"grad_norm": 7.333261966705322,
|
| 2604 |
+
"learning_rate": 8.953214733288383e-06,
|
| 2605 |
+
"loss": 2.4145,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.8532110091743119,
|
| 2610 |
+
"grad_norm": 7.81040096282959,
|
| 2611 |
+
"learning_rate": 8.688061284200266e-06,
|
| 2612 |
+
"loss": 2.5398,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.8555045871559633,
|
| 2617 |
+
"grad_norm": 7.140593528747559,
|
| 2618 |
+
"learning_rate": 8.426519384872733e-06,
|
| 2619 |
+
"loss": 1.8465,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.8577981651376146,
|
| 2624 |
+
"grad_norm": 8.926170349121094,
|
| 2625 |
+
"learning_rate": 8.168611899679013e-06,
|
| 2626 |
+
"loss": 2.4094,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8600917431192661,
|
| 2631 |
+
"grad_norm": 8.442564010620117,
|
| 2632 |
+
"learning_rate": 7.914361375266504e-06,
|
| 2633 |
+
"loss": 2.2949,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.8623853211009175,
|
| 2638 |
+
"grad_norm": 7.919826030731201,
|
| 2639 |
+
"learning_rate": 7.663790038585793e-06,
|
| 2640 |
+
"loss": 2.1068,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.8646788990825688,
|
| 2645 |
+
"grad_norm": 8.824353218078613,
|
| 2646 |
+
"learning_rate": 7.416919794947536e-06,
|
| 2647 |
+
"loss": 2.7126,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.8669724770642202,
|
| 2652 |
+
"grad_norm": 9.781707763671875,
|
| 2653 |
+
"learning_rate": 7.173772226107434e-06,
|
| 2654 |
+
"loss": 2.4447,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.8692660550458715,
|
| 2659 |
+
"grad_norm": 6.505077838897705,
|
| 2660 |
+
"learning_rate": 6.934368588379553e-06,
|
| 2661 |
+
"loss": 1.1717,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.8715596330275229,
|
| 2666 |
+
"grad_norm": 9.621846199035645,
|
| 2667 |
+
"learning_rate": 6.698729810778065e-06,
|
| 2668 |
+
"loss": 1.8266,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.8738532110091743,
|
| 2673 |
+
"grad_norm": 9.959877014160156,
|
| 2674 |
+
"learning_rate": 6.46687649318759e-06,
|
| 2675 |
+
"loss": 1.2395,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.8761467889908257,
|
| 2680 |
+
"grad_norm": 5.53802490234375,
|
| 2681 |
+
"learning_rate": 6.238828904562316e-06,
|
| 2682 |
+
"loss": 0.4439,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.8784403669724771,
|
| 2687 |
+
"grad_norm": 4.555595397949219,
|
| 2688 |
+
"learning_rate": 6.014606981154086e-06,
|
| 2689 |
+
"loss": 0.3088,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.8807339449541285,
|
| 2694 |
+
"grad_norm": 9.899633407592773,
|
| 2695 |
+
"learning_rate": 5.794230324769517e-06,
|
| 2696 |
+
"loss": 1.7023,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.8830275229357798,
|
| 2701 |
+
"grad_norm": 11.428282737731934,
|
| 2702 |
+
"learning_rate": 5.577718201056392e-06,
|
| 2703 |
+
"loss": 2.7028,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.8853211009174312,
|
| 2708 |
+
"grad_norm": 11.778576850891113,
|
| 2709 |
+
"learning_rate": 5.365089537819434e-06,
|
| 2710 |
+
"loss": 2.2125,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.8876146788990825,
|
| 2715 |
+
"grad_norm": 13.558716773986816,
|
| 2716 |
+
"learning_rate": 5.156362923365588e-06,
|
| 2717 |
+
"loss": 2.5294,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.8899082568807339,
|
| 2722 |
+
"grad_norm": 14.217622756958008,
|
| 2723 |
+
"learning_rate": 4.951556604879048e-06,
|
| 2724 |
+
"loss": 2.5287,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.8922018348623854,
|
| 2729 |
+
"grad_norm": 10.065847396850586,
|
| 2730 |
+
"learning_rate": 4.7506884868259995e-06,
|
| 2731 |
+
"loss": 1.123,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.8944954128440367,
|
| 2736 |
+
"grad_norm": 45.029380798339844,
|
| 2737 |
+
"learning_rate": 4.5537761293894535e-06,
|
| 2738 |
+
"loss": 0.8917,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.8967889908256881,
|
| 2743 |
+
"grad_norm": 14.9079008102417,
|
| 2744 |
+
"learning_rate": 4.360836746934055e-06,
|
| 2745 |
+
"loss": 0.6098,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.8990825688073395,
|
| 2750 |
+
"grad_norm": 3.1857004165649414,
|
| 2751 |
+
"learning_rate": 4.1718872065011904e-06,
|
| 2752 |
+
"loss": 0.4474,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.9013761467889908,
|
| 2757 |
+
"grad_norm": 3.916781425476074,
|
| 2758 |
+
"learning_rate": 3.9869440263344714e-06,
|
| 2759 |
+
"loss": 0.6104,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.9036697247706422,
|
| 2764 |
+
"grad_norm": 3.7978272438049316,
|
| 2765 |
+
"learning_rate": 3.8060233744356633e-06,
|
| 2766 |
+
"loss": 0.4462,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.9059633027522935,
|
| 2771 |
+
"grad_norm": 4.325674057006836,
|
| 2772 |
+
"learning_rate": 3.6291410671512594e-06,
|
| 2773 |
+
"loss": 0.467,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.908256880733945,
|
| 2778 |
+
"grad_norm": 3.895176887512207,
|
| 2779 |
+
"learning_rate": 3.4563125677897932e-06,
|
| 2780 |
+
"loss": 0.4774,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.9105504587155964,
|
| 2785 |
+
"grad_norm": 4.118838787078857,
|
| 2786 |
+
"learning_rate": 3.2875529852700147e-06,
|
| 2787 |
+
"loss": 0.6108,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.9128440366972477,
|
| 2792 |
+
"grad_norm": 3.1684885025024414,
|
| 2793 |
+
"learning_rate": 3.1228770728000455e-06,
|
| 2794 |
+
"loss": 0.5086,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.9151376146788991,
|
| 2799 |
+
"grad_norm": 4.82421350479126,
|
| 2800 |
+
"learning_rate": 2.9622992265876392e-06,
|
| 2801 |
+
"loss": 0.7931,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.9174311926605505,
|
| 2806 |
+
"grad_norm": 4.262173652648926,
|
| 2807 |
+
"learning_rate": 2.8058334845816213e-06,
|
| 2808 |
+
"loss": 0.7849,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.9197247706422018,
|
| 2813 |
+
"grad_norm": 1.1381596326828003,
|
| 2814 |
+
"learning_rate": 2.653493525244721e-06,
|
| 2815 |
+
"loss": 0.6932,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.9220183486238532,
|
| 2820 |
+
"grad_norm": 1.1581947803497314,
|
| 2821 |
+
"learning_rate": 2.5052926663577e-06,
|
| 2822 |
+
"loss": 0.5253,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.9243119266055045,
|
| 2827 |
+
"grad_norm": 1.5183253288269043,
|
| 2828 |
+
"learning_rate": 2.361243863855184e-06,
|
| 2829 |
+
"loss": 0.853,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.926605504587156,
|
| 2834 |
+
"grad_norm": 1.571021556854248,
|
| 2835 |
+
"learning_rate": 2.221359710692961e-06,
|
| 2836 |
+
"loss": 0.6197,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.9288990825688074,
|
| 2841 |
+
"grad_norm": 1.6061910390853882,
|
| 2842 |
+
"learning_rate": 2.085652435747132e-06,
|
| 2843 |
+
"loss": 0.7312,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.9311926605504587,
|
| 2848 |
+
"grad_norm": 2.495539665222168,
|
| 2849 |
+
"learning_rate": 1.9541339027450256e-06,
|
| 2850 |
+
"loss": 1.4046,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.9334862385321101,
|
| 2855 |
+
"grad_norm": 2.363898277282715,
|
| 2856 |
+
"learning_rate": 1.8268156092280496e-06,
|
| 2857 |
+
"loss": 1.1194,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.9357798165137615,
|
| 2862 |
+
"grad_norm": 2.802872657775879,
|
| 2863 |
+
"learning_rate": 1.70370868554659e-06,
|
| 2864 |
+
"loss": 1.3099,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.9380733944954128,
|
| 2869 |
+
"grad_norm": 3.056849479675293,
|
| 2870 |
+
"learning_rate": 1.584823893886933e-06,
|
| 2871 |
+
"loss": 1.1909,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.9403669724770642,
|
| 2876 |
+
"grad_norm": 3.2700366973876953,
|
| 2877 |
+
"learning_rate": 1.4701716273304521e-06,
|
| 2878 |
+
"loss": 1.2238,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.9426605504587156,
|
| 2883 |
+
"grad_norm": 3.4624550342559814,
|
| 2884 |
+
"learning_rate": 1.3597619089450342e-06,
|
| 2885 |
+
"loss": 1.2718,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.944954128440367,
|
| 2890 |
+
"grad_norm": 4.333258628845215,
|
| 2891 |
+
"learning_rate": 1.2536043909088191e-06,
|
| 2892 |
+
"loss": 1.7671,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.9472477064220184,
|
| 2897 |
+
"grad_norm": 4.528647422790527,
|
| 2898 |
+
"learning_rate": 1.1517083536664142e-06,
|
| 2899 |
+
"loss": 1.4721,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.9495412844036697,
|
| 2904 |
+
"grad_norm": 5.667140483856201,
|
| 2905 |
+
"learning_rate": 1.0540827051175818e-06,
|
| 2906 |
+
"loss": 1.8333,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.9518348623853211,
|
| 2911 |
+
"grad_norm": 7.816887378692627,
|
| 2912 |
+
"learning_rate": 9.607359798384785e-07,
|
| 2913 |
+
"loss": 2.4681,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.9541284403669725,
|
| 2918 |
+
"grad_norm": 6.785305976867676,
|
| 2919 |
+
"learning_rate": 8.716763383355864e-07,
|
| 2920 |
+
"loss": 2.1718,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.9564220183486238,
|
| 2925 |
+
"grad_norm": 5.145172595977783,
|
| 2926 |
+
"learning_rate": 7.869115663322879e-07,
|
| 2927 |
+
"loss": 1.2761,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.9587155963302753,
|
| 2932 |
+
"grad_norm": 5.58955717086792,
|
| 2933 |
+
"learning_rate": 7.064490740882057e-07,
|
| 2934 |
+
"loss": 1.5643,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.9610091743119266,
|
| 2939 |
+
"grad_norm": 7.9990458488464355,
|
| 2940 |
+
"learning_rate": 6.302958957514371e-07,
|
| 2941 |
+
"loss": 1.9113,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.963302752293578,
|
| 2946 |
+
"grad_norm": 8.559749603271484,
|
| 2947 |
+
"learning_rate": 5.584586887435739e-07,
|
| 2948 |
+
"loss": 2.0603,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.9655963302752294,
|
| 2953 |
+
"grad_norm": 8.403243064880371,
|
| 2954 |
+
"learning_rate": 4.909437331777179e-07,
|
| 2955 |
+
"loss": 2.0897,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.9678899082568807,
|
| 2960 |
+
"grad_norm": 8.703038215637207,
|
| 2961 |
+
"learning_rate": 4.277569313094809e-07,
|
| 2962 |
+
"loss": 1.4529,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.9701834862385321,
|
| 2967 |
+
"grad_norm": 10.667278289794922,
|
| 2968 |
+
"learning_rate": 3.689038070209594e-07,
|
| 2969 |
+
"loss": 1.466,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.9724770642201835,
|
| 2974 |
+
"grad_norm": 8.07142162322998,
|
| 2975 |
+
"learning_rate": 3.143895053378698e-07,
|
| 2976 |
+
"loss": 1.4229,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.9747706422018348,
|
| 2981 |
+
"grad_norm": 6.67749547958374,
|
| 2982 |
+
"learning_rate": 2.6421879197974784e-07,
|
| 2983 |
+
"loss": 0.8897,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.9770642201834863,
|
| 2988 |
+
"grad_norm": 10.8878755569458,
|
| 2989 |
+
"learning_rate": 2.1839605294330933e-07,
|
| 2990 |
+
"loss": 2.0611,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.9793577981651376,
|
| 2995 |
+
"grad_norm": 13.075303077697754,
|
| 2996 |
+
"learning_rate": 1.7692529411904578e-07,
|
| 2997 |
+
"loss": 2.3857,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.981651376146789,
|
| 3002 |
+
"grad_norm": 16.536373138427734,
|
| 3003 |
+
"learning_rate": 1.3981014094099353e-07,
|
| 3004 |
+
"loss": 2.6068,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.9839449541284404,
|
| 3009 |
+
"grad_norm": 20.786617279052734,
|
| 3010 |
+
"learning_rate": 1.0705383806982606e-07,
|
| 3011 |
+
"loss": 1.1238,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.9862385321100917,
|
| 3016 |
+
"grad_norm": 12.918766975402832,
|
| 3017 |
+
"learning_rate": 7.865924910916977e-08,
|
| 3018 |
+
"loss": 0.4258,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.9885321100917431,
|
| 3023 |
+
"grad_norm": 3.077303647994995,
|
| 3024 |
+
"learning_rate": 5.462885635529324e-08,
|
| 3025 |
+
"loss": 0.5306,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.9908256880733946,
|
| 3030 |
+
"grad_norm": 3.1448254585266113,
|
| 3031 |
+
"learning_rate": 3.496476058006959e-08,
|
| 3032 |
+
"loss": 0.4716,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.9931192660550459,
|
| 3037 |
+
"grad_norm": 3.913860321044922,
|
| 3038 |
+
"learning_rate": 1.9668680847356735e-08,
|
| 3039 |
+
"loss": 0.4871,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.9954128440366973,
|
| 3044 |
+
"grad_norm": 3.7353949546813965,
|
| 3045 |
+
"learning_rate": 8.741954362678772e-09,
|
| 3046 |
+
"loss": 0.5689,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.9977064220183486,
|
| 3051 |
+
"grad_norm": 3.645770788192749,
|
| 3052 |
+
"learning_rate": 2.185536356363871e-09,
|
| 3053 |
+
"loss": 0.6416,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 1.0,
|
| 3058 |
+
"grad_norm": 4.725067615509033,
|
| 3059 |
+
"learning_rate": 0.0,
|
| 3060 |
+
"loss": 0.7461,
|
| 3061 |
+
"step": 436
|
| 3062 |
}
|
| 3063 |
],
|
| 3064 |
"logging_steps": 1,
|
|
|
|
| 3073 |
"should_evaluate": false,
|
| 3074 |
"should_log": false,
|
| 3075 |
"should_save": true,
|
| 3076 |
+
"should_training_stop": true
|
| 3077 |
},
|
| 3078 |
"attributes": {}
|
| 3079 |
}
|
| 3080 |
},
|
| 3081 |
+
"total_flos": 2.0262676808073216e+16,
|
| 3082 |
"train_batch_size": 4,
|
| 3083 |
"trial_name": null,
|
| 3084 |
"trial_params": null
|