sft-6700-step
Browse files- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +1338 -6
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 698419728
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5afe2a027d395f551dd07fb95d989917901b4fc4d9a3f699bcf96a1a0c52045c
|
| 3 |
size 698419728
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1397136587
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc2f55384d96aef8c36f040c4c76bba8c886457259f6de979d38405706b46bfd
|
| 3 |
size 1397136587
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14645
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14f4ff77bbbe606f2785ea0f04a2535a7c54901e7091c0fed26a6e7b85eda9ae
|
| 3 |
size 14645
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08804d0a21d8df191a267fdfad60532380afcf3b04182506e89486e7b012afc5
|
| 3 |
size 1465
|
trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 1.
|
| 4 |
-
"best_model_checkpoint": "/workspace/project_2026_1/checkpoints/sft/checkpoint-
|
| 5 |
-
"epoch": 1.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -6113,6 +6113,1338 @@
|
|
| 6113 |
"eval_samples_per_second": 26.036,
|
| 6114 |
"eval_steps_per_second": 3.258,
|
| 6115 |
"step": 5500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6116 |
}
|
| 6117 |
],
|
| 6118 |
"logging_steps": 10,
|
|
@@ -6132,7 +7464,7 @@
|
|
| 6132 |
"attributes": {}
|
| 6133 |
}
|
| 6134 |
},
|
| 6135 |
-
"total_flos": 1.
|
| 6136 |
"train_batch_size": 8,
|
| 6137 |
"trial_name": null,
|
| 6138 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 6700,
|
| 3 |
+
"best_metric": 1.140816330909729,
|
| 4 |
+
"best_model_checkpoint": "/workspace/project_2026_1/checkpoints/sft/checkpoint-6700",
|
| 5 |
+
"epoch": 1.9711679905854664,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 6700,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 6113 |
"eval_samples_per_second": 26.036,
|
| 6114 |
"eval_steps_per_second": 3.258,
|
| 6115 |
"step": 5500
|
| 6116 |
+
},
|
| 6117 |
+
{
|
| 6118 |
+
"entropy": 1.0032637119293213,
|
| 6119 |
+
"epoch": 1.6210650191232716,
|
| 6120 |
+
"grad_norm": 0.5667787194252014,
|
| 6121 |
+
"learning_rate": 9.49591141020415e-05,
|
| 6122 |
+
"loss": 1.0222906112670898,
|
| 6123 |
+
"mean_token_accuracy": 0.7597410917282105,
|
| 6124 |
+
"num_tokens": 22252809.0,
|
| 6125 |
+
"step": 5510
|
| 6126 |
+
},
|
| 6127 |
+
{
|
| 6128 |
+
"entropy": 0.9571436047554016,
|
| 6129 |
+
"epoch": 1.6240070609002648,
|
| 6130 |
+
"grad_norm": 0.5619379281997681,
|
| 6131 |
+
"learning_rate": 9.463524331205183e-05,
|
| 6132 |
+
"loss": 0.9508034706115722,
|
| 6133 |
+
"mean_token_accuracy": 0.7767503380775451,
|
| 6134 |
+
"num_tokens": 22293323.0,
|
| 6135 |
+
"step": 5520
|
| 6136 |
+
},
|
| 6137 |
+
{
|
| 6138 |
+
"entropy": 0.9551607072353363,
|
| 6139 |
+
"epoch": 1.626949102677258,
|
| 6140 |
+
"grad_norm": 0.5588138699531555,
|
| 6141 |
+
"learning_rate": 9.431142894696174e-05,
|
| 6142 |
+
"loss": 0.968760871887207,
|
| 6143 |
+
"mean_token_accuracy": 0.7735034108161927,
|
| 6144 |
+
"num_tokens": 22333734.0,
|
| 6145 |
+
"step": 5530
|
| 6146 |
+
},
|
| 6147 |
+
{
|
| 6148 |
+
"entropy": 0.9952653288841248,
|
| 6149 |
+
"epoch": 1.6298911444542512,
|
| 6150 |
+
"grad_norm": 0.6699135899543762,
|
| 6151 |
+
"learning_rate": 9.398767441255356e-05,
|
| 6152 |
+
"loss": 1.0053104400634765,
|
| 6153 |
+
"mean_token_accuracy": 0.7669021725654602,
|
| 6154 |
+
"num_tokens": 22373932.0,
|
| 6155 |
+
"step": 5540
|
| 6156 |
+
},
|
| 6157 |
+
{
|
| 6158 |
+
"entropy": 0.9081286668777466,
|
| 6159 |
+
"epoch": 1.6328331862312444,
|
| 6160 |
+
"grad_norm": 0.5072507858276367,
|
| 6161 |
+
"learning_rate": 9.366398311398013e-05,
|
| 6162 |
+
"loss": 0.9016472816467285,
|
| 6163 |
+
"mean_token_accuracy": 0.7893808305263519,
|
| 6164 |
+
"num_tokens": 22413965.0,
|
| 6165 |
+
"step": 5550
|
| 6166 |
+
},
|
| 6167 |
+
{
|
| 6168 |
+
"entropy": 0.8828806400299072,
|
| 6169 |
+
"epoch": 1.6357752280082378,
|
| 6170 |
+
"grad_norm": 0.5978274941444397,
|
| 6171 |
+
"learning_rate": 9.334035845572945e-05,
|
| 6172 |
+
"loss": 0.8982778549194336,
|
| 6173 |
+
"mean_token_accuracy": 0.7933094263076782,
|
| 6174 |
+
"num_tokens": 22454470.0,
|
| 6175 |
+
"step": 5560
|
| 6176 |
+
},
|
| 6177 |
+
{
|
| 6178 |
+
"entropy": 0.9773084461688996,
|
| 6179 |
+
"epoch": 1.6387172697852308,
|
| 6180 |
+
"grad_norm": 0.6699267029762268,
|
| 6181 |
+
"learning_rate": 9.301680384158834e-05,
|
| 6182 |
+
"loss": 0.9834060668945312,
|
| 6183 |
+
"mean_token_accuracy": 0.7695248782634735,
|
| 6184 |
+
"num_tokens": 22494739.0,
|
| 6185 |
+
"step": 5570
|
| 6186 |
+
},
|
| 6187 |
+
{
|
| 6188 |
+
"entropy": 0.9948910295963287,
|
| 6189 |
+
"epoch": 1.6416593115622242,
|
| 6190 |
+
"grad_norm": 0.6252007484436035,
|
| 6191 |
+
"learning_rate": 9.269332267460717e-05,
|
| 6192 |
+
"loss": 1.0045485496520996,
|
| 6193 |
+
"mean_token_accuracy": 0.7613544285297393,
|
| 6194 |
+
"num_tokens": 22535199.0,
|
| 6195 |
+
"step": 5580
|
| 6196 |
+
},
|
| 6197 |
+
{
|
| 6198 |
+
"entropy": 0.9052659869194031,
|
| 6199 |
+
"epoch": 1.6446013533392174,
|
| 6200 |
+
"grad_norm": 0.6294770240783691,
|
| 6201 |
+
"learning_rate": 9.236991835706361e-05,
|
| 6202 |
+
"loss": 0.9096663475036622,
|
| 6203 |
+
"mean_token_accuracy": 0.7889019668102264,
|
| 6204 |
+
"num_tokens": 22575473.0,
|
| 6205 |
+
"step": 5590
|
| 6206 |
+
},
|
| 6207 |
+
{
|
| 6208 |
+
"entropy": 0.9944833815097809,
|
| 6209 |
+
"epoch": 1.6475433951162106,
|
| 6210 |
+
"grad_norm": 0.6281090378761292,
|
| 6211 |
+
"learning_rate": 9.204659429042723e-05,
|
| 6212 |
+
"loss": 1.001138401031494,
|
| 6213 |
+
"mean_token_accuracy": 0.766866946220398,
|
| 6214 |
+
"num_tokens": 22615909.0,
|
| 6215 |
+
"step": 5600
|
| 6216 |
+
},
|
| 6217 |
+
{
|
| 6218 |
+
"epoch": 1.6475433951162106,
|
| 6219 |
+
"eval_entropy": 1.0232390931897901,
|
| 6220 |
+
"eval_loss": 1.1652106046676636,
|
| 6221 |
+
"eval_mean_token_accuracy": 0.7458901661900398,
|
| 6222 |
+
"eval_num_tokens": 22615909.0,
|
| 6223 |
+
"eval_runtime": 116.9138,
|
| 6224 |
+
"eval_samples_per_second": 26.045,
|
| 6225 |
+
"eval_steps_per_second": 3.259,
|
| 6226 |
+
"step": 5600
|
| 6227 |
+
},
|
| 6228 |
+
{
|
| 6229 |
+
"entropy": 0.9154108762741089,
|
| 6230 |
+
"epoch": 1.650485436893204,
|
| 6231 |
+
"grad_norm": 0.6548126935958862,
|
| 6232 |
+
"learning_rate": 9.172335387532337e-05,
|
| 6233 |
+
"loss": 0.9116446495056152,
|
| 6234 |
+
"mean_token_accuracy": 0.7843240320682525,
|
| 6235 |
+
"num_tokens": 22656406.0,
|
| 6236 |
+
"step": 5610
|
| 6237 |
+
},
|
| 6238 |
+
{
|
| 6239 |
+
"entropy": 0.9418833196163178,
|
| 6240 |
+
"epoch": 1.653427478670197,
|
| 6241 |
+
"grad_norm": 0.5818053483963013,
|
| 6242 |
+
"learning_rate": 9.140020051149768e-05,
|
| 6243 |
+
"loss": 0.9640005111694336,
|
| 6244 |
+
"mean_token_accuracy": 0.7761074662208557,
|
| 6245 |
+
"num_tokens": 22697053.0,
|
| 6246 |
+
"step": 5620
|
| 6247 |
+
},
|
| 6248 |
+
{
|
| 6249 |
+
"entropy": 1.0457957625389098,
|
| 6250 |
+
"epoch": 1.6563695204471904,
|
| 6251 |
+
"grad_norm": 0.563709020614624,
|
| 6252 |
+
"learning_rate": 9.107713759778022e-05,
|
| 6253 |
+
"loss": 1.0556187629699707,
|
| 6254 |
+
"mean_token_accuracy": 0.7564215302467346,
|
| 6255 |
+
"num_tokens": 22737587.0,
|
| 6256 |
+
"step": 5630
|
| 6257 |
+
},
|
| 6258 |
+
{
|
| 6259 |
+
"entropy": 0.9574743151664734,
|
| 6260 |
+
"epoch": 1.6593115622241836,
|
| 6261 |
+
"grad_norm": 0.5461353659629822,
|
| 6262 |
+
"learning_rate": 9.075416853204962e-05,
|
| 6263 |
+
"loss": 0.9538365364074707,
|
| 6264 |
+
"mean_token_accuracy": 0.7788529932498932,
|
| 6265 |
+
"num_tokens": 22778087.0,
|
| 6266 |
+
"step": 5640
|
| 6267 |
+
},
|
| 6268 |
+
{
|
| 6269 |
+
"entropy": 0.9407840669155121,
|
| 6270 |
+
"epoch": 1.6622536040011768,
|
| 6271 |
+
"grad_norm": 0.6229255199432373,
|
| 6272 |
+
"learning_rate": 9.043129671119757e-05,
|
| 6273 |
+
"loss": 0.9592649459838867,
|
| 6274 |
+
"mean_token_accuracy": 0.7759726524353028,
|
| 6275 |
+
"num_tokens": 22818636.0,
|
| 6276 |
+
"step": 5650
|
| 6277 |
+
},
|
| 6278 |
+
{
|
| 6279 |
+
"entropy": 0.9525330364704132,
|
| 6280 |
+
"epoch": 1.66519564577817,
|
| 6281 |
+
"grad_norm": 0.517737627029419,
|
| 6282 |
+
"learning_rate": 9.010852553109286e-05,
|
| 6283 |
+
"loss": 0.9527083396911621,
|
| 6284 |
+
"mean_token_accuracy": 0.7764707028865814,
|
| 6285 |
+
"num_tokens": 22858762.0,
|
| 6286 |
+
"step": 5660
|
| 6287 |
+
},
|
| 6288 |
+
{
|
| 6289 |
+
"entropy": 0.9349599361419678,
|
| 6290 |
+
"epoch": 1.6681376875551632,
|
| 6291 |
+
"grad_norm": 0.5596241354942322,
|
| 6292 |
+
"learning_rate": 8.97858583865459e-05,
|
| 6293 |
+
"loss": 0.958860206604004,
|
| 6294 |
+
"mean_token_accuracy": 0.7790948987007141,
|
| 6295 |
+
"num_tokens": 22898952.0,
|
| 6296 |
+
"step": 5670
|
| 6297 |
+
},
|
| 6298 |
+
{
|
| 6299 |
+
"entropy": 1.007050359249115,
|
| 6300 |
+
"epoch": 1.6710797293321566,
|
| 6301 |
+
"grad_norm": 0.5459507703781128,
|
| 6302 |
+
"learning_rate": 8.946329867127271e-05,
|
| 6303 |
+
"loss": 1.009850311279297,
|
| 6304 |
+
"mean_token_accuracy": 0.7651616752147674,
|
| 6305 |
+
"num_tokens": 22939561.0,
|
| 6306 |
+
"step": 5680
|
| 6307 |
+
},
|
| 6308 |
+
{
|
| 6309 |
+
"entropy": 1.0099753439426422,
|
| 6310 |
+
"epoch": 1.6740217711091496,
|
| 6311 |
+
"grad_norm": 0.5861520171165466,
|
| 6312 |
+
"learning_rate": 8.91408497778596e-05,
|
| 6313 |
+
"loss": 1.0120088577270507,
|
| 6314 |
+
"mean_token_accuracy": 0.7637844383716583,
|
| 6315 |
+
"num_tokens": 22979861.0,
|
| 6316 |
+
"step": 5690
|
| 6317 |
+
},
|
| 6318 |
+
{
|
| 6319 |
+
"entropy": 0.9966277837753296,
|
| 6320 |
+
"epoch": 1.676963812886143,
|
| 6321 |
+
"grad_norm": 0.5946571826934814,
|
| 6322 |
+
"learning_rate": 8.88185150977271e-05,
|
| 6323 |
+
"loss": 1.007601833343506,
|
| 6324 |
+
"mean_token_accuracy": 0.7666476130485534,
|
| 6325 |
+
"num_tokens": 23020082.0,
|
| 6326 |
+
"step": 5700
|
| 6327 |
+
},
|
| 6328 |
+
{
|
| 6329 |
+
"epoch": 1.676963812886143,
|
| 6330 |
+
"eval_entropy": 1.0183745707113911,
|
| 6331 |
+
"eval_loss": 1.1620737314224243,
|
| 6332 |
+
"eval_mean_token_accuracy": 0.7465485564367039,
|
| 6333 |
+
"eval_num_tokens": 23020082.0,
|
| 6334 |
+
"eval_runtime": 116.931,
|
| 6335 |
+
"eval_samples_per_second": 26.041,
|
| 6336 |
+
"eval_steps_per_second": 3.258,
|
| 6337 |
+
"step": 5700
|
| 6338 |
+
},
|
| 6339 |
+
{
|
| 6340 |
+
"entropy": 0.9435779750347137,
|
| 6341 |
+
"epoch": 1.6799058546631362,
|
| 6342 |
+
"grad_norm": 0.6357449293136597,
|
| 6343 |
+
"learning_rate": 8.849629802109463e-05,
|
| 6344 |
+
"loss": 0.9518097877502442,
|
| 6345 |
+
"mean_token_accuracy": 0.7776012420654297,
|
| 6346 |
+
"num_tokens": 23060478.0,
|
| 6347 |
+
"step": 5710
|
| 6348 |
+
},
|
| 6349 |
+
{
|
| 6350 |
+
"entropy": 0.9298249185085297,
|
| 6351 |
+
"epoch": 1.6828478964401294,
|
| 6352 |
+
"grad_norm": 0.5645864009857178,
|
| 6353 |
+
"learning_rate": 8.817420193694458e-05,
|
| 6354 |
+
"loss": 0.9362122535705566,
|
| 6355 |
+
"mean_token_accuracy": 0.7816990375518799,
|
| 6356 |
+
"num_tokens": 23100989.0,
|
| 6357 |
+
"step": 5720
|
| 6358 |
+
},
|
| 6359 |
+
{
|
| 6360 |
+
"entropy": 0.9536093771457672,
|
| 6361 |
+
"epoch": 1.6857899382171226,
|
| 6362 |
+
"grad_norm": 0.6313226819038391,
|
| 6363 |
+
"learning_rate": 8.785223023298684e-05,
|
| 6364 |
+
"loss": 0.9790238380432129,
|
| 6365 |
+
"mean_token_accuracy": 0.7684462070465088,
|
| 6366 |
+
"num_tokens": 23141372.0,
|
| 6367 |
+
"step": 5730
|
| 6368 |
+
},
|
| 6369 |
+
{
|
| 6370 |
+
"entropy": 0.980869197845459,
|
| 6371 |
+
"epoch": 1.6887319799941158,
|
| 6372 |
+
"grad_norm": 0.5673984885215759,
|
| 6373 |
+
"learning_rate": 8.753038629562313e-05,
|
| 6374 |
+
"loss": 0.9640357971191407,
|
| 6375 |
+
"mean_token_accuracy": 0.7741734623908997,
|
| 6376 |
+
"num_tokens": 23181824.0,
|
| 6377 |
+
"step": 5740
|
| 6378 |
+
},
|
| 6379 |
+
{
|
| 6380 |
+
"entropy": 0.9453027725219727,
|
| 6381 |
+
"epoch": 1.6916740217711093,
|
| 6382 |
+
"grad_norm": 0.6107286214828491,
|
| 6383 |
+
"learning_rate": 8.720867350991128e-05,
|
| 6384 |
+
"loss": 0.973298454284668,
|
| 6385 |
+
"mean_token_accuracy": 0.7727717578411102,
|
| 6386 |
+
"num_tokens": 23222185.0,
|
| 6387 |
+
"step": 5750
|
| 6388 |
+
},
|
| 6389 |
+
{
|
| 6390 |
+
"entropy": 0.9626853942871094,
|
| 6391 |
+
"epoch": 1.6946160635481022,
|
| 6392 |
+
"grad_norm": 0.5757405757904053,
|
| 6393 |
+
"learning_rate": 8.688709525952978e-05,
|
| 6394 |
+
"loss": 0.9758338928222656,
|
| 6395 |
+
"mean_token_accuracy": 0.772016829252243,
|
| 6396 |
+
"num_tokens": 23262748.0,
|
| 6397 |
+
"step": 5760
|
| 6398 |
+
},
|
| 6399 |
+
{
|
| 6400 |
+
"entropy": 1.010034316778183,
|
| 6401 |
+
"epoch": 1.6975581053250957,
|
| 6402 |
+
"grad_norm": 0.5179459452629089,
|
| 6403 |
+
"learning_rate": 8.656565492674205e-05,
|
| 6404 |
+
"loss": 1.0052608489990233,
|
| 6405 |
+
"mean_token_accuracy": 0.7618889510631561,
|
| 6406 |
+
"num_tokens": 23303156.0,
|
| 6407 |
+
"step": 5770
|
| 6408 |
+
},
|
| 6409 |
+
{
|
| 6410 |
+
"entropy": 0.8848201990127563,
|
| 6411 |
+
"epoch": 1.7005001471020889,
|
| 6412 |
+
"grad_norm": 0.5171420574188232,
|
| 6413 |
+
"learning_rate": 8.6244355892361e-05,
|
| 6414 |
+
"loss": 0.8950881958007812,
|
| 6415 |
+
"mean_token_accuracy": 0.7918769180774688,
|
| 6416 |
+
"num_tokens": 23343542.0,
|
| 6417 |
+
"step": 5780
|
| 6418 |
+
},
|
| 6419 |
+
{
|
| 6420 |
+
"entropy": 0.97591233253479,
|
| 6421 |
+
"epoch": 1.703442188879082,
|
| 6422 |
+
"grad_norm": 0.5939028859138489,
|
| 6423 |
+
"learning_rate": 8.592320153571334e-05,
|
| 6424 |
+
"loss": 1.006781005859375,
|
| 6425 |
+
"mean_token_accuracy": 0.7655499637126922,
|
| 6426 |
+
"num_tokens": 23384111.0,
|
| 6427 |
+
"step": 5790
|
| 6428 |
+
},
|
| 6429 |
+
{
|
| 6430 |
+
"entropy": 1.0557245433330535,
|
| 6431 |
+
"epoch": 1.7063842306560755,
|
| 6432 |
+
"grad_norm": 0.5849112868309021,
|
| 6433 |
+
"learning_rate": 8.56021952346042e-05,
|
| 6434 |
+
"loss": 1.0547979354858399,
|
| 6435 |
+
"mean_token_accuracy": 0.7541541934013367,
|
| 6436 |
+
"num_tokens": 23424325.0,
|
| 6437 |
+
"step": 5800
|
| 6438 |
+
},
|
| 6439 |
+
{
|
| 6440 |
+
"epoch": 1.7063842306560755,
|
| 6441 |
+
"eval_entropy": 1.0359160511356014,
|
| 6442 |
+
"eval_loss": 1.1574015617370605,
|
| 6443 |
+
"eval_mean_token_accuracy": 0.7470338016044437,
|
| 6444 |
+
"eval_num_tokens": 23424325.0,
|
| 6445 |
+
"eval_runtime": 117.0424,
|
| 6446 |
+
"eval_samples_per_second": 26.016,
|
| 6447 |
+
"eval_steps_per_second": 3.255,
|
| 6448 |
+
"step": 5800
|
| 6449 |
+
},
|
| 6450 |
+
{
|
| 6451 |
+
"entropy": 0.9237102508544922,
|
| 6452 |
+
"epoch": 1.7093262724330684,
|
| 6453 |
+
"grad_norm": 0.6681418418884277,
|
| 6454 |
+
"learning_rate": 8.528134036528137e-05,
|
| 6455 |
+
"loss": 0.9113170623779296,
|
| 6456 |
+
"mean_token_accuracy": 0.7863486051559448,
|
| 6457 |
+
"num_tokens": 23464980.0,
|
| 6458 |
+
"step": 5810
|
| 6459 |
+
},
|
| 6460 |
+
{
|
| 6461 |
+
"entropy": 0.9206311643123627,
|
| 6462 |
+
"epoch": 1.7122683142100619,
|
| 6463 |
+
"grad_norm": 0.6511064767837524,
|
| 6464 |
+
"learning_rate": 8.496064030240009e-05,
|
| 6465 |
+
"loss": 0.9329397201538085,
|
| 6466 |
+
"mean_token_accuracy": 0.7813911080360413,
|
| 6467 |
+
"num_tokens": 23505413.0,
|
| 6468 |
+
"step": 5820
|
| 6469 |
+
},
|
| 6470 |
+
{
|
| 6471 |
+
"entropy": 0.9037490963935852,
|
| 6472 |
+
"epoch": 1.715210355987055,
|
| 6473 |
+
"grad_norm": 0.5818105936050415,
|
| 6474 |
+
"learning_rate": 8.464009841898721e-05,
|
| 6475 |
+
"loss": 0.9231014251708984,
|
| 6476 |
+
"mean_token_accuracy": 0.7818919003009797,
|
| 6477 |
+
"num_tokens": 23545900.0,
|
| 6478 |
+
"step": 5830
|
| 6479 |
+
},
|
| 6480 |
+
{
|
| 6481 |
+
"entropy": 0.9442632138729096,
|
| 6482 |
+
"epoch": 1.7181523977640483,
|
| 6483 |
+
"grad_norm": 0.5522459745407104,
|
| 6484 |
+
"learning_rate": 8.431971808640604e-05,
|
| 6485 |
+
"loss": 0.9421136856079102,
|
| 6486 |
+
"mean_token_accuracy": 0.7791707038879394,
|
| 6487 |
+
"num_tokens": 23586448.0,
|
| 6488 |
+
"step": 5840
|
| 6489 |
+
},
|
| 6490 |
+
{
|
| 6491 |
+
"entropy": 0.9893321812152862,
|
| 6492 |
+
"epoch": 1.7210944395410415,
|
| 6493 |
+
"grad_norm": 0.5959654450416565,
|
| 6494 |
+
"learning_rate": 8.399950267432073e-05,
|
| 6495 |
+
"loss": 0.9956209182739257,
|
| 6496 |
+
"mean_token_accuracy": 0.7700819849967957,
|
| 6497 |
+
"num_tokens": 23626702.0,
|
| 6498 |
+
"step": 5850
|
| 6499 |
+
},
|
| 6500 |
+
{
|
| 6501 |
+
"entropy": 0.9568989872932434,
|
| 6502 |
+
"epoch": 1.7240364813180347,
|
| 6503 |
+
"grad_norm": 0.573197066783905,
|
| 6504 |
+
"learning_rate": 8.367945555066072e-05,
|
| 6505 |
+
"loss": 0.9633212089538574,
|
| 6506 |
+
"mean_token_accuracy": 0.7744130253791809,
|
| 6507 |
+
"num_tokens": 23666990.0,
|
| 6508 |
+
"step": 5860
|
| 6509 |
+
},
|
| 6510 |
+
{
|
| 6511 |
+
"entropy": 0.9612068772315979,
|
| 6512 |
+
"epoch": 1.726978523095028,
|
| 6513 |
+
"grad_norm": 0.5556803941726685,
|
| 6514 |
+
"learning_rate": 8.335958008158564e-05,
|
| 6515 |
+
"loss": 0.9782578468322753,
|
| 6516 |
+
"mean_token_accuracy": 0.7722434878349305,
|
| 6517 |
+
"num_tokens": 23707419.0,
|
| 6518 |
+
"step": 5870
|
| 6519 |
+
},
|
| 6520 |
+
{
|
| 6521 |
+
"entropy": 0.9662620186805725,
|
| 6522 |
+
"epoch": 1.729920564872021,
|
| 6523 |
+
"grad_norm": 0.5682553648948669,
|
| 6524 |
+
"learning_rate": 8.303987963144951e-05,
|
| 6525 |
+
"loss": 0.980229663848877,
|
| 6526 |
+
"mean_token_accuracy": 0.7722213745117188,
|
| 6527 |
+
"num_tokens": 23747899.0,
|
| 6528 |
+
"step": 5880
|
| 6529 |
+
},
|
| 6530 |
+
{
|
| 6531 |
+
"entropy": 0.9563367903232575,
|
| 6532 |
+
"epoch": 1.7328626066490145,
|
| 6533 |
+
"grad_norm": 0.644250214099884,
|
| 6534 |
+
"learning_rate": 8.272035756276574e-05,
|
| 6535 |
+
"loss": 0.9506181716918946,
|
| 6536 |
+
"mean_token_accuracy": 0.7787576794624329,
|
| 6537 |
+
"num_tokens": 23788539.0,
|
| 6538 |
+
"step": 5890
|
| 6539 |
+
},
|
| 6540 |
+
{
|
| 6541 |
+
"entropy": 0.9529013156890869,
|
| 6542 |
+
"epoch": 1.7358046484260077,
|
| 6543 |
+
"grad_norm": 0.6191350817680359,
|
| 6544 |
+
"learning_rate": 8.240101723617138e-05,
|
| 6545 |
+
"loss": 0.9677048683166504,
|
| 6546 |
+
"mean_token_accuracy": 0.7735692262649536,
|
| 6547 |
+
"num_tokens": 23829048.0,
|
| 6548 |
+
"step": 5900
|
| 6549 |
+
},
|
| 6550 |
+
{
|
| 6551 |
+
"epoch": 1.7358046484260077,
|
| 6552 |
+
"eval_entropy": 1.0162569234377443,
|
| 6553 |
+
"eval_loss": 1.158705234527588,
|
| 6554 |
+
"eval_mean_token_accuracy": 0.7472113214452748,
|
| 6555 |
+
"eval_num_tokens": 23829048.0,
|
| 6556 |
+
"eval_runtime": 117.138,
|
| 6557 |
+
"eval_samples_per_second": 25.995,
|
| 6558 |
+
"eval_steps_per_second": 3.253,
|
| 6559 |
+
"step": 5900
|
| 6560 |
+
},
|
| 6561 |
+
{
|
| 6562 |
+
"entropy": 0.9016938626766204,
|
| 6563 |
+
"epoch": 1.7387466902030009,
|
| 6564 |
+
"grad_norm": 0.5551770329475403,
|
| 6565 |
+
"learning_rate": 8.208186201039215e-05,
|
| 6566 |
+
"loss": 0.9020340919494629,
|
| 6567 |
+
"mean_token_accuracy": 0.7864646017551422,
|
| 6568 |
+
"num_tokens": 23869463.0,
|
| 6569 |
+
"step": 5910
|
| 6570 |
+
},
|
| 6571 |
+
{
|
| 6572 |
+
"entropy": 0.9148910164833068,
|
| 6573 |
+
"epoch": 1.741688731979994,
|
| 6574 |
+
"grad_norm": 0.6301241517066956,
|
| 6575 |
+
"learning_rate": 8.176289524220682e-05,
|
| 6576 |
+
"loss": 0.9415826797485352,
|
| 6577 |
+
"mean_token_accuracy": 0.7839051008224487,
|
| 6578 |
+
"num_tokens": 23909844.0,
|
| 6579 |
+
"step": 5920
|
| 6580 |
+
},
|
| 6581 |
+
{
|
| 6582 |
+
"entropy": 0.9921420514583588,
|
| 6583 |
+
"epoch": 1.7446307737569873,
|
| 6584 |
+
"grad_norm": 0.5697426199913025,
|
| 6585 |
+
"learning_rate": 8.144412028641213e-05,
|
| 6586 |
+
"loss": 0.9879927635192871,
|
| 6587 |
+
"mean_token_accuracy": 0.7715160071849823,
|
| 6588 |
+
"num_tokens": 23949594.0,
|
| 6589 |
+
"step": 5930
|
| 6590 |
+
},
|
| 6591 |
+
{
|
| 6592 |
+
"entropy": 0.9673746526241302,
|
| 6593 |
+
"epoch": 1.7475728155339807,
|
| 6594 |
+
"grad_norm": 0.7032338976860046,
|
| 6595 |
+
"learning_rate": 8.112554049578723e-05,
|
| 6596 |
+
"loss": 0.9581163406372071,
|
| 6597 |
+
"mean_token_accuracy": 0.7709846436977387,
|
| 6598 |
+
"num_tokens": 23989917.0,
|
| 6599 |
+
"step": 5940
|
| 6600 |
+
},
|
| 6601 |
+
{
|
| 6602 |
+
"entropy": 0.9318383634090424,
|
| 6603 |
+
"epoch": 1.7505148573109737,
|
| 6604 |
+
"grad_norm": 0.5314655900001526,
|
| 6605 |
+
"learning_rate": 8.080715922105873e-05,
|
| 6606 |
+
"loss": 0.948590087890625,
|
| 6607 |
+
"mean_token_accuracy": 0.7803104817867279,
|
| 6608 |
+
"num_tokens": 24029981.0,
|
| 6609 |
+
"step": 5950
|
| 6610 |
+
},
|
| 6611 |
+
{
|
| 6612 |
+
"entropy": 0.9275396764278412,
|
| 6613 |
+
"epoch": 1.753456899087967,
|
| 6614 |
+
"grad_norm": 0.6153765320777893,
|
| 6615 |
+
"learning_rate": 8.048897981086527e-05,
|
| 6616 |
+
"loss": 0.9455156326293945,
|
| 6617 |
+
"mean_token_accuracy": 0.7832433462142945,
|
| 6618 |
+
"num_tokens": 24070430.0,
|
| 6619 |
+
"step": 5960
|
| 6620 |
+
},
|
| 6621 |
+
{
|
| 6622 |
+
"entropy": 0.9975035011768341,
|
| 6623 |
+
"epoch": 1.7563989408649603,
|
| 6624 |
+
"grad_norm": 0.5570023059844971,
|
| 6625 |
+
"learning_rate": 8.017100561172228e-05,
|
| 6626 |
+
"loss": 0.9986433982849121,
|
| 6627 |
+
"mean_token_accuracy": 0.7639374434947968,
|
| 6628 |
+
"num_tokens": 24110808.0,
|
| 6629 |
+
"step": 5970
|
| 6630 |
+
},
|
| 6631 |
+
{
|
| 6632 |
+
"entropy": 0.985156637430191,
|
| 6633 |
+
"epoch": 1.7593409826419535,
|
| 6634 |
+
"grad_norm": 0.5818830132484436,
|
| 6635 |
+
"learning_rate": 7.985323996798696e-05,
|
| 6636 |
+
"loss": 1.0052236557006835,
|
| 6637 |
+
"mean_token_accuracy": 0.7657383203506469,
|
| 6638 |
+
"num_tokens": 24151389.0,
|
| 6639 |
+
"step": 5980
|
| 6640 |
+
},
|
| 6641 |
+
{
|
| 6642 |
+
"entropy": 0.8605277717113495,
|
| 6643 |
+
"epoch": 1.762283024418947,
|
| 6644 |
+
"grad_norm": 0.6189702153205872,
|
| 6645 |
+
"learning_rate": 7.953568622182283e-05,
|
| 6646 |
+
"loss": 0.8629916191101075,
|
| 6647 |
+
"mean_token_accuracy": 0.8023348867893219,
|
| 6648 |
+
"num_tokens": 24191459.0,
|
| 6649 |
+
"step": 5990
|
| 6650 |
+
},
|
| 6651 |
+
{
|
| 6652 |
+
"entropy": 0.9501297771930695,
|
| 6653 |
+
"epoch": 1.7652250661959399,
|
| 6654 |
+
"grad_norm": 0.6018420457839966,
|
| 6655 |
+
"learning_rate": 7.921834771316489e-05,
|
| 6656 |
+
"loss": 0.9521394729614258,
|
| 6657 |
+
"mean_token_accuracy": 0.7792097210884095,
|
| 6658 |
+
"num_tokens": 24231918.0,
|
| 6659 |
+
"step": 6000
|
| 6660 |
+
},
|
| 6661 |
+
{
|
| 6662 |
+
"epoch": 1.7652250661959399,
|
| 6663 |
+
"eval_entropy": 1.0099175068650033,
|
| 6664 |
+
"eval_loss": 1.1563514471054077,
|
| 6665 |
+
"eval_mean_token_accuracy": 0.7476834809686256,
|
| 6666 |
+
"eval_num_tokens": 24231918.0,
|
| 6667 |
+
"eval_runtime": 117.1325,
|
| 6668 |
+
"eval_samples_per_second": 25.996,
|
| 6669 |
+
"eval_steps_per_second": 3.253,
|
| 6670 |
+
"step": 6000
|
| 6671 |
+
},
|
| 6672 |
+
{
|
| 6673 |
+
"entropy": 0.9893746197223663,
|
| 6674 |
+
"epoch": 1.7681671079729333,
|
| 6675 |
+
"grad_norm": 0.6696788668632507,
|
| 6676 |
+
"learning_rate": 7.890122777968419e-05,
|
| 6677 |
+
"loss": 1.0002012252807617,
|
| 6678 |
+
"mean_token_accuracy": 0.7695098519325256,
|
| 6679 |
+
"num_tokens": 24271949.0,
|
| 6680 |
+
"step": 6010
|
| 6681 |
+
},
|
| 6682 |
+
{
|
| 6683 |
+
"entropy": 0.9130432605743408,
|
| 6684 |
+
"epoch": 1.7711091497499265,
|
| 6685 |
+
"grad_norm": 0.558464527130127,
|
| 6686 |
+
"learning_rate": 7.858432975675304e-05,
|
| 6687 |
+
"loss": 0.9236593246459961,
|
| 6688 |
+
"mean_token_accuracy": 0.7876148581504822,
|
| 6689 |
+
"num_tokens": 24312102.0,
|
| 6690 |
+
"step": 6020
|
| 6691 |
+
},
|
| 6692 |
+
{
|
| 6693 |
+
"entropy": 0.9415574371814728,
|
| 6694 |
+
"epoch": 1.7740511915269197,
|
| 6695 |
+
"grad_norm": 0.5865418314933777,
|
| 6696 |
+
"learning_rate": 7.826765697740957e-05,
|
| 6697 |
+
"loss": 0.9479406356811524,
|
| 6698 |
+
"mean_token_accuracy": 0.7780888140201568,
|
| 6699 |
+
"num_tokens": 24352693.0,
|
| 6700 |
+
"step": 6030
|
| 6701 |
+
},
|
| 6702 |
+
{
|
| 6703 |
+
"entropy": 0.9924417555332183,
|
| 6704 |
+
"epoch": 1.776993233303913,
|
| 6705 |
+
"grad_norm": 0.6885764598846436,
|
| 6706 |
+
"learning_rate": 7.795121277232302e-05,
|
| 6707 |
+
"loss": 0.9945806503295899,
|
| 6708 |
+
"mean_token_accuracy": 0.7648349285125733,
|
| 6709 |
+
"num_tokens": 24393351.0,
|
| 6710 |
+
"step": 6040
|
| 6711 |
+
},
|
| 6712 |
+
{
|
| 6713 |
+
"entropy": 0.984746390581131,
|
| 6714 |
+
"epoch": 1.779935275080906,
|
| 6715 |
+
"grad_norm": 0.5965508222579956,
|
| 6716 |
+
"learning_rate": 7.763500046975853e-05,
|
| 6717 |
+
"loss": 0.9933752059936524,
|
| 6718 |
+
"mean_token_accuracy": 0.7661060273647309,
|
| 6719 |
+
"num_tokens": 24433759.0,
|
| 6720 |
+
"step": 6050
|
| 6721 |
+
},
|
| 6722 |
+
{
|
| 6723 |
+
"entropy": 0.9540082156658173,
|
| 6724 |
+
"epoch": 1.7828773168578995,
|
| 6725 |
+
"grad_norm": 0.5714673399925232,
|
| 6726 |
+
"learning_rate": 7.731902339554206e-05,
|
| 6727 |
+
"loss": 0.9627056121826172,
|
| 6728 |
+
"mean_token_accuracy": 0.7745068728923797,
|
| 6729 |
+
"num_tokens": 24474130.0,
|
| 6730 |
+
"step": 6060
|
| 6731 |
+
},
|
| 6732 |
+
{
|
| 6733 |
+
"entropy": 0.9777018666267395,
|
| 6734 |
+
"epoch": 1.7858193586348925,
|
| 6735 |
+
"grad_norm": 0.6372447609901428,
|
| 6736 |
+
"learning_rate": 7.700328487302565e-05,
|
| 6737 |
+
"loss": 1.0012830734252929,
|
| 6738 |
+
"mean_token_accuracy": 0.7722647070884705,
|
| 6739 |
+
"num_tokens": 24514543.0,
|
| 6740 |
+
"step": 6070
|
| 6741 |
+
},
|
| 6742 |
+
{
|
| 6743 |
+
"entropy": 0.9325755536556244,
|
| 6744 |
+
"epoch": 1.788761400411886,
|
| 6745 |
+
"grad_norm": 0.6036056876182556,
|
| 6746 |
+
"learning_rate": 7.668778822305217e-05,
|
| 6747 |
+
"loss": 0.9412235260009766,
|
| 6748 |
+
"mean_token_accuracy": 0.7827513933181762,
|
| 6749 |
+
"num_tokens": 24554910.0,
|
| 6750 |
+
"step": 6080
|
| 6751 |
+
},
|
| 6752 |
+
{
|
| 6753 |
+
"entropy": 0.9325757265090943,
|
| 6754 |
+
"epoch": 1.7917034421888791,
|
| 6755 |
+
"grad_norm": 0.5404115319252014,
|
| 6756 |
+
"learning_rate": 7.637253676392074e-05,
|
| 6757 |
+
"loss": 0.9362624168395997,
|
| 6758 |
+
"mean_token_accuracy": 0.7818540692329407,
|
| 6759 |
+
"num_tokens": 24595471.0,
|
| 6760 |
+
"step": 6090
|
| 6761 |
+
},
|
| 6762 |
+
{
|
| 6763 |
+
"entropy": 0.9863007128238678,
|
| 6764 |
+
"epoch": 1.7946454839658723,
|
| 6765 |
+
"grad_norm": 0.5063208341598511,
|
| 6766 |
+
"learning_rate": 7.605753381135138e-05,
|
| 6767 |
+
"loss": 0.9955570220947265,
|
| 6768 |
+
"mean_token_accuracy": 0.7685826182365417,
|
| 6769 |
+
"num_tokens": 24635954.0,
|
| 6770 |
+
"step": 6100
|
| 6771 |
+
},
|
| 6772 |
+
{
|
| 6773 |
+
"epoch": 1.7946454839658723,
|
| 6774 |
+
"eval_entropy": 1.018375001085086,
|
| 6775 |
+
"eval_loss": 1.1510425806045532,
|
| 6776 |
+
"eval_mean_token_accuracy": 0.7484834164459249,
|
| 6777 |
+
"eval_num_tokens": 24635954.0,
|
| 6778 |
+
"eval_runtime": 117.0613,
|
| 6779 |
+
"eval_samples_per_second": 26.012,
|
| 6780 |
+
"eval_steps_per_second": 3.255,
|
| 6781 |
+
"step": 6100
|
| 6782 |
+
},
|
| 6783 |
+
{
|
| 6784 |
+
"entropy": 0.9858646035194397,
|
| 6785 |
+
"epoch": 1.7975875257428655,
|
| 6786 |
+
"grad_norm": 0.5223714113235474,
|
| 6787 |
+
"learning_rate": 7.574278267845069e-05,
|
| 6788 |
+
"loss": 0.9794650077819824,
|
| 6789 |
+
"mean_token_accuracy": 0.7720021724700927,
|
| 6790 |
+
"num_tokens": 24676082.0,
|
| 6791 |
+
"step": 6110
|
| 6792 |
+
},
|
| 6793 |
+
{
|
| 6794 |
+
"entropy": 0.9158162832260132,
|
| 6795 |
+
"epoch": 1.8005295675198587,
|
| 6796 |
+
"grad_norm": 0.5589065551757812,
|
| 6797 |
+
"learning_rate": 7.542828667567643e-05,
|
| 6798 |
+
"loss": 0.9300893783569336,
|
| 6799 |
+
"mean_token_accuracy": 0.7815610766410828,
|
| 6800 |
+
"num_tokens": 24716092.0,
|
| 6801 |
+
"step": 6120
|
| 6802 |
+
},
|
| 6803 |
+
{
|
| 6804 |
+
"entropy": 0.9701406300067902,
|
| 6805 |
+
"epoch": 1.8034716092968521,
|
| 6806 |
+
"grad_norm": 0.5639879107475281,
|
| 6807 |
+
"learning_rate": 7.511404911080319e-05,
|
| 6808 |
+
"loss": 0.9756647109985351,
|
| 6809 |
+
"mean_token_accuracy": 0.7717652976512909,
|
| 6810 |
+
"num_tokens": 24756492.0,
|
| 6811 |
+
"step": 6130
|
| 6812 |
+
},
|
| 6813 |
+
{
|
| 6814 |
+
"entropy": 0.9247495532035828,
|
| 6815 |
+
"epoch": 1.806413651073845,
|
| 6816 |
+
"grad_norm": 0.5797274708747864,
|
| 6817 |
+
"learning_rate": 7.480007328888724e-05,
|
| 6818 |
+
"loss": 0.936155128479004,
|
| 6819 |
+
"mean_token_accuracy": 0.7829030692577362,
|
| 6820 |
+
"num_tokens": 24796749.0,
|
| 6821 |
+
"step": 6140
|
| 6822 |
+
},
|
| 6823 |
+
{
|
| 6824 |
+
"entropy": 0.938025814294815,
|
| 6825 |
+
"epoch": 1.8093556928508385,
|
| 6826 |
+
"grad_norm": 0.5666895508766174,
|
| 6827 |
+
"learning_rate": 7.4486362512232e-05,
|
| 6828 |
+
"loss": 0.9645057678222656,
|
| 6829 |
+
"mean_token_accuracy": 0.7746810853481293,
|
| 6830 |
+
"num_tokens": 24837331.0,
|
| 6831 |
+
"step": 6150
|
| 6832 |
+
},
|
| 6833 |
+
{
|
| 6834 |
+
"entropy": 0.9641852080821991,
|
| 6835 |
+
"epoch": 1.8122977346278317,
|
| 6836 |
+
"grad_norm": 0.599879801273346,
|
| 6837 |
+
"learning_rate": 7.417292008035324e-05,
|
| 6838 |
+
"loss": 0.9777825355529786,
|
| 6839 |
+
"mean_token_accuracy": 0.7724472403526306,
|
| 6840 |
+
"num_tokens": 24877763.0,
|
| 6841 |
+
"step": 6160
|
| 6842 |
+
},
|
| 6843 |
+
{
|
| 6844 |
+
"entropy": 0.9553499221801758,
|
| 6845 |
+
"epoch": 1.815239776404825,
|
| 6846 |
+
"grad_norm": 0.6117326617240906,
|
| 6847 |
+
"learning_rate": 7.385974928994424e-05,
|
| 6848 |
+
"loss": 0.9453524589538574,
|
| 6849 |
+
"mean_token_accuracy": 0.7802985668182373,
|
| 6850 |
+
"num_tokens": 24918137.0,
|
| 6851 |
+
"step": 6170
|
| 6852 |
+
},
|
| 6853 |
+
{
|
| 6854 |
+
"entropy": 0.9422242820262909,
|
| 6855 |
+
"epoch": 1.8181818181818183,
|
| 6856 |
+
"grad_norm": 0.5855090618133545,
|
| 6857 |
+
"learning_rate": 7.354685343484141e-05,
|
| 6858 |
+
"loss": 0.9547875404357911,
|
| 6859 |
+
"mean_token_accuracy": 0.7761439323425293,
|
| 6860 |
+
"num_tokens": 24958540.0,
|
| 6861 |
+
"step": 6180
|
| 6862 |
+
},
|
| 6863 |
+
{
|
| 6864 |
+
"entropy": 0.9967733740806579,
|
| 6865 |
+
"epoch": 1.8211238599588113,
|
| 6866 |
+
"grad_norm": 0.6166797876358032,
|
| 6867 |
+
"learning_rate": 7.323423580598929e-05,
|
| 6868 |
+
"loss": 1.0093032836914062,
|
| 6869 |
+
"mean_token_accuracy": 0.7644744515419006,
|
| 6870 |
+
"num_tokens": 24999010.0,
|
| 6871 |
+
"step": 6190
|
| 6872 |
+
},
|
| 6873 |
+
{
|
| 6874 |
+
"entropy": 0.941829913854599,
|
| 6875 |
+
"epoch": 1.8240659017358047,
|
| 6876 |
+
"grad_norm": 0.5976388454437256,
|
| 6877 |
+
"learning_rate": 7.292189969140627e-05,
|
| 6878 |
+
"loss": 0.947477912902832,
|
| 6879 |
+
"mean_token_accuracy": 0.7809899151325226,
|
| 6880 |
+
"num_tokens": 25039340.0,
|
| 6881 |
+
"step": 6200
|
| 6882 |
+
},
|
| 6883 |
+
{
|
| 6884 |
+
"epoch": 1.8240659017358047,
|
| 6885 |
+
"eval_entropy": 1.0210820323056748,
|
| 6886 |
+
"eval_loss": 1.149355173110962,
|
| 6887 |
+
"eval_mean_token_accuracy": 0.7488088197908377,
|
| 6888 |
+
"eval_num_tokens": 25039340.0,
|
| 6889 |
+
"eval_runtime": 117.0315,
|
| 6890 |
+
"eval_samples_per_second": 26.019,
|
| 6891 |
+
"eval_steps_per_second": 3.256,
|
| 6892 |
+
"step": 6200
|
| 6893 |
+
},
|
| 6894 |
+
{
|
| 6895 |
+
"entropy": 0.968848192691803,
|
| 6896 |
+
"epoch": 1.8270079435127977,
|
| 6897 |
+
"grad_norm": 0.5740299224853516,
|
| 6898 |
+
"learning_rate": 7.260984837614976e-05,
|
| 6899 |
+
"loss": 0.97001953125,
|
| 6900 |
+
"mean_token_accuracy": 0.7758668541908265,
|
| 6901 |
+
"num_tokens": 25079782.0,
|
| 6902 |
+
"step": 6210
|
| 6903 |
+
},
|
| 6904 |
+
{
|
| 6905 |
+
"entropy": 0.9311902105808259,
|
| 6906 |
+
"epoch": 1.8299499852897911,
|
| 6907 |
+
"grad_norm": 0.595485508441925,
|
| 6908 |
+
"learning_rate": 7.229808514228182e-05,
|
| 6909 |
+
"loss": 0.9422932624816894,
|
| 6910 |
+
"mean_token_accuracy": 0.779161137342453,
|
| 6911 |
+
"num_tokens": 25120387.0,
|
| 6912 |
+
"step": 6220
|
| 6913 |
+
},
|
| 6914 |
+
{
|
| 6915 |
+
"entropy": 0.904720538854599,
|
| 6916 |
+
"epoch": 1.8328920270667843,
|
| 6917 |
+
"grad_norm": 0.6421226263046265,
|
| 6918 |
+
"learning_rate": 7.198661326883446e-05,
|
| 6919 |
+
"loss": 0.8994368553161621,
|
| 6920 |
+
"mean_token_accuracy": 0.7901805222034455,
|
| 6921 |
+
"num_tokens": 25160952.0,
|
| 6922 |
+
"step": 6230
|
| 6923 |
+
},
|
| 6924 |
+
{
|
| 6925 |
+
"entropy": 0.9412755012512207,
|
| 6926 |
+
"epoch": 1.8358340688437775,
|
| 6927 |
+
"grad_norm": 0.5241137742996216,
|
| 6928 |
+
"learning_rate": 7.167543603177535e-05,
|
| 6929 |
+
"loss": 0.9635884284973144,
|
| 6930 |
+
"mean_token_accuracy": 0.7794827520847321,
|
| 6931 |
+
"num_tokens": 25201298.0,
|
| 6932 |
+
"step": 6240
|
| 6933 |
+
},
|
| 6934 |
+
{
|
| 6935 |
+
"entropy": 0.9426091015338898,
|
| 6936 |
+
"epoch": 1.838776110620771,
|
| 6937 |
+
"grad_norm": 0.599162220954895,
|
| 6938 |
+
"learning_rate": 7.136455670397317e-05,
|
| 6939 |
+
"loss": 0.9373735427856446,
|
| 6940 |
+
"mean_token_accuracy": 0.7805384159088135,
|
| 6941 |
+
"num_tokens": 25241730.0,
|
| 6942 |
+
"step": 6250
|
| 6943 |
+
},
|
| 6944 |
+
{
|
| 6945 |
+
"entropy": 0.8965267837047577,
|
| 6946 |
+
"epoch": 1.841718152397764,
|
| 6947 |
+
"grad_norm": 0.5050822496414185,
|
| 6948 |
+
"learning_rate": 7.105397855516332e-05,
|
| 6949 |
+
"loss": 0.8982341766357422,
|
| 6950 |
+
"mean_token_accuracy": 0.7888839960098266,
|
| 6951 |
+
"num_tokens": 25282116.0,
|
| 6952 |
+
"step": 6260
|
| 6953 |
+
},
|
| 6954 |
+
{
|
| 6955 |
+
"entropy": 0.9214460253715515,
|
| 6956 |
+
"epoch": 1.8446601941747574,
|
| 6957 |
+
"grad_norm": 0.6280463933944702,
|
| 6958 |
+
"learning_rate": 7.074370485191353e-05,
|
| 6959 |
+
"loss": 0.9190691947937012,
|
| 6960 |
+
"mean_token_accuracy": 0.7820965766906738,
|
| 6961 |
+
"num_tokens": 25322627.0,
|
| 6962 |
+
"step": 6270
|
| 6963 |
+
},
|
| 6964 |
+
{
|
| 6965 |
+
"entropy": 0.892383873462677,
|
| 6966 |
+
"epoch": 1.8476022359517505,
|
| 6967 |
+
"grad_norm": 0.582876980304718,
|
| 6968 |
+
"learning_rate": 7.043373885758938e-05,
|
| 6969 |
+
"loss": 0.9149553298950195,
|
| 6970 |
+
"mean_token_accuracy": 0.786136794090271,
|
| 6971 |
+
"num_tokens": 25362914.0,
|
| 6972 |
+
"step": 6280
|
| 6973 |
+
},
|
| 6974 |
+
{
|
| 6975 |
+
"entropy": 0.9371729671955109,
|
| 6976 |
+
"epoch": 1.8505442777287437,
|
| 6977 |
+
"grad_norm": 0.5959920287132263,
|
| 6978 |
+
"learning_rate": 7.012408383232016e-05,
|
| 6979 |
+
"loss": 0.9573171615600586,
|
| 6980 |
+
"mean_token_accuracy": 0.7788842916488647,
|
| 6981 |
+
"num_tokens": 25403259.0,
|
| 6982 |
+
"step": 6290
|
| 6983 |
+
},
|
| 6984 |
+
{
|
| 6985 |
+
"entropy": 0.9701938092708587,
|
| 6986 |
+
"epoch": 1.853486319505737,
|
| 6987 |
+
"grad_norm": 0.5748452544212341,
|
| 6988 |
+
"learning_rate": 6.981474303296436e-05,
|
| 6989 |
+
"loss": 0.9730587005615234,
|
| 6990 |
+
"mean_token_accuracy": 0.7726157248020172,
|
| 6991 |
+
"num_tokens": 25443719.0,
|
| 6992 |
+
"step": 6300
|
| 6993 |
+
},
|
| 6994 |
+
{
|
| 6995 |
+
"epoch": 1.853486319505737,
|
| 6996 |
+
"eval_entropy": 1.0170868068855266,
|
| 6997 |
+
"eval_loss": 1.148207426071167,
|
| 6998 |
+
"eval_mean_token_accuracy": 0.7491307286765632,
|
| 6999 |
+
"eval_num_tokens": 25443719.0,
|
| 7000 |
+
"eval_runtime": 116.8997,
|
| 7001 |
+
"eval_samples_per_second": 26.048,
|
| 7002 |
+
"eval_steps_per_second": 3.259,
|
| 7003 |
+
"step": 6300
|
| 7004 |
+
},
|
| 7005 |
+
{
|
| 7006 |
+
"entropy": 0.9635331034660339,
|
| 7007 |
+
"epoch": 1.8564283612827301,
|
| 7008 |
+
"grad_norm": 0.6185274720191956,
|
| 7009 |
+
"learning_rate": 6.950571971307566e-05,
|
| 7010 |
+
"loss": 0.9592901229858398,
|
| 7011 |
+
"mean_token_accuracy": 0.7760994493961334,
|
| 7012 |
+
"num_tokens": 25484099.0,
|
| 7013 |
+
"step": 6310
|
| 7014 |
+
},
|
| 7015 |
+
{
|
| 7016 |
+
"entropy": 0.9286525547504425,
|
| 7017 |
+
"epoch": 1.8593704030597236,
|
| 7018 |
+
"grad_norm": 0.6655700206756592,
|
| 7019 |
+
"learning_rate": 6.919701712286845e-05,
|
| 7020 |
+
"loss": 0.949979591369629,
|
| 7021 |
+
"mean_token_accuracy": 0.7771933794021606,
|
| 7022 |
+
"num_tokens": 25524367.0,
|
| 7023 |
+
"step": 6320
|
| 7024 |
+
},
|
| 7025 |
+
{
|
| 7026 |
+
"entropy": 0.9677168548107147,
|
| 7027 |
+
"epoch": 1.8623124448367165,
|
| 7028 |
+
"grad_norm": 0.6009240746498108,
|
| 7029 |
+
"learning_rate": 6.888863850918397e-05,
|
| 7030 |
+
"loss": 0.9582759857177734,
|
| 7031 |
+
"mean_token_accuracy": 0.7760764300823212,
|
| 7032 |
+
"num_tokens": 25564686.0,
|
| 7033 |
+
"step": 6330
|
| 7034 |
+
},
|
| 7035 |
+
{
|
| 7036 |
+
"entropy": 0.8910984754562378,
|
| 7037 |
+
"epoch": 1.86525448661371,
|
| 7038 |
+
"grad_norm": 0.5529386401176453,
|
| 7039 |
+
"learning_rate": 6.858058711545576e-05,
|
| 7040 |
+
"loss": 0.8990293502807617,
|
| 7041 |
+
"mean_token_accuracy": 0.7910106480121613,
|
| 7042 |
+
"num_tokens": 25604757.0,
|
| 7043 |
+
"step": 6340
|
| 7044 |
+
},
|
| 7045 |
+
{
|
| 7046 |
+
"entropy": 0.9290744543075562,
|
| 7047 |
+
"epoch": 1.8681965283907032,
|
| 7048 |
+
"grad_norm": 0.5405545234680176,
|
| 7049 |
+
"learning_rate": 6.827286618167593e-05,
|
| 7050 |
+
"loss": 0.9487281799316406,
|
| 7051 |
+
"mean_token_accuracy": 0.7793277740478516,
|
| 7052 |
+
"num_tokens": 25645415.0,
|
| 7053 |
+
"step": 6350
|
| 7054 |
+
},
|
| 7055 |
+
{
|
| 7056 |
+
"entropy": 0.9454367399215698,
|
| 7057 |
+
"epoch": 1.8711385701676964,
|
| 7058 |
+
"grad_norm": 0.5304380059242249,
|
| 7059 |
+
"learning_rate": 6.796547894436078e-05,
|
| 7060 |
+
"loss": 0.9380789756774902,
|
| 7061 |
+
"mean_token_accuracy": 0.7802990317344666,
|
| 7062 |
+
"num_tokens": 25685916.0,
|
| 7063 |
+
"step": 6360
|
| 7064 |
+
},
|
| 7065 |
+
{
|
| 7066 |
+
"entropy": 0.9380621254444123,
|
| 7067 |
+
"epoch": 1.8740806119446898,
|
| 7068 |
+
"grad_norm": 0.5882839560508728,
|
| 7069 |
+
"learning_rate": 6.7658428636517e-05,
|
| 7070 |
+
"loss": 0.9496076583862305,
|
| 7071 |
+
"mean_token_accuracy": 0.7798330843448639,
|
| 7072 |
+
"num_tokens": 25726346.0,
|
| 7073 |
+
"step": 6370
|
| 7074 |
+
},
|
| 7075 |
+
{
|
| 7076 |
+
"entropy": 0.9862012684345245,
|
| 7077 |
+
"epoch": 1.8770226537216828,
|
| 7078 |
+
"grad_norm": 0.614160418510437,
|
| 7079 |
+
"learning_rate": 6.735171848760753e-05,
|
| 7080 |
+
"loss": 1.0059632301330566,
|
| 7081 |
+
"mean_token_accuracy": 0.7649614214897156,
|
| 7082 |
+
"num_tokens": 25766824.0,
|
| 7083 |
+
"step": 6380
|
| 7084 |
+
},
|
| 7085 |
+
{
|
| 7086 |
+
"entropy": 0.9311308979988098,
|
| 7087 |
+
"epoch": 1.8799646954986762,
|
| 7088 |
+
"grad_norm": 0.6380969882011414,
|
| 7089 |
+
"learning_rate": 6.704535172351752e-05,
|
| 7090 |
+
"loss": 0.9215089797973632,
|
| 7091 |
+
"mean_token_accuracy": 0.7852351903915405,
|
| 7092 |
+
"num_tokens": 25807230.0,
|
| 7093 |
+
"step": 6390
|
| 7094 |
+
},
|
| 7095 |
+
{
|
| 7096 |
+
"entropy": 0.9507179379463195,
|
| 7097 |
+
"epoch": 1.8829067372756692,
|
| 7098 |
+
"grad_norm": 0.5307073593139648,
|
| 7099 |
+
"learning_rate": 6.673933156652068e-05,
|
| 7100 |
+
"loss": 0.9578986167907715,
|
| 7101 |
+
"mean_token_accuracy": 0.7749875366687775,
|
| 7102 |
+
"num_tokens": 25847836.0,
|
| 7103 |
+
"step": 6400
|
| 7104 |
+
},
|
| 7105 |
+
{
|
| 7106 |
+
"epoch": 1.8829067372756692,
|
| 7107 |
+
"eval_entropy": 0.996097157007753,
|
| 7108 |
+
"eval_loss": 1.1488333940505981,
|
| 7109 |
+
"eval_mean_token_accuracy": 0.7494633900211864,
|
| 7110 |
+
"eval_num_tokens": 25847836.0,
|
| 7111 |
+
"eval_runtime": 116.9155,
|
| 7112 |
+
"eval_samples_per_second": 26.044,
|
| 7113 |
+
"eval_steps_per_second": 3.259,
|
| 7114 |
+
"step": 6400
|
| 7115 |
+
},
|
| 7116 |
+
{
|
| 7117 |
+
"entropy": 0.9198467969894409,
|
| 7118 |
+
"epoch": 1.8858487790526626,
|
| 7119 |
+
"grad_norm": 0.4932635426521301,
|
| 7120 |
+
"learning_rate": 6.643366123524502e-05,
|
| 7121 |
+
"loss": 0.9418664932250976,
|
| 7122 |
+
"mean_token_accuracy": 0.7808252394199371,
|
| 7123 |
+
"num_tokens": 25887974.0,
|
| 7124 |
+
"step": 6410
|
| 7125 |
+
},
|
| 7126 |
+
{
|
| 7127 |
+
"entropy": 0.9422998249530792,
|
| 7128 |
+
"epoch": 1.8887908208296558,
|
| 7129 |
+
"grad_norm": 0.5207474231719971,
|
| 7130 |
+
"learning_rate": 6.612834394463936e-05,
|
| 7131 |
+
"loss": 0.9341909408569335,
|
| 7132 |
+
"mean_token_accuracy": 0.7823355376720429,
|
| 7133 |
+
"num_tokens": 25927559.0,
|
| 7134 |
+
"step": 6420
|
| 7135 |
+
},
|
| 7136 |
+
{
|
| 7137 |
+
"entropy": 1.0029336094856263,
|
| 7138 |
+
"epoch": 1.891732862606649,
|
| 7139 |
+
"grad_norm": 0.6492148041725159,
|
| 7140 |
+
"learning_rate": 6.582338290593918e-05,
|
| 7141 |
+
"loss": 1.0240444183349608,
|
| 7142 |
+
"mean_token_accuracy": 0.7613063991069794,
|
| 7143 |
+
"num_tokens": 25968038.0,
|
| 7144 |
+
"step": 6430
|
| 7145 |
+
},
|
| 7146 |
+
{
|
| 7147 |
+
"entropy": 0.9833552062511444,
|
| 7148 |
+
"epoch": 1.8946749043836424,
|
| 7149 |
+
"grad_norm": 0.590535044670105,
|
| 7150 |
+
"learning_rate": 6.55187813266332e-05,
|
| 7151 |
+
"loss": 1.003129005432129,
|
| 7152 |
+
"mean_token_accuracy": 0.7678163826465607,
|
| 7153 |
+
"num_tokens": 26008348.0,
|
| 7154 |
+
"step": 6440
|
| 7155 |
+
},
|
| 7156 |
+
{
|
| 7157 |
+
"entropy": 0.8829509735107421,
|
| 7158 |
+
"epoch": 1.8976169461606354,
|
| 7159 |
+
"grad_norm": 0.574521005153656,
|
| 7160 |
+
"learning_rate": 6.521454241042924e-05,
|
| 7161 |
+
"loss": 0.8665548324584961,
|
| 7162 |
+
"mean_token_accuracy": 0.7962438046932221,
|
| 7163 |
+
"num_tokens": 26048371.0,
|
| 7164 |
+
"step": 6450
|
| 7165 |
+
},
|
| 7166 |
+
{
|
| 7167 |
+
"entropy": 0.9392775595188141,
|
| 7168 |
+
"epoch": 1.9005589879376288,
|
| 7169 |
+
"grad_norm": 0.5355552434921265,
|
| 7170 |
+
"learning_rate": 6.491066935722091e-05,
|
| 7171 |
+
"loss": 0.9620846748352051,
|
| 7172 |
+
"mean_token_accuracy": 0.7788772523403168,
|
| 7173 |
+
"num_tokens": 26088711.0,
|
| 7174 |
+
"step": 6460
|
| 7175 |
+
},
|
| 7176 |
+
{
|
| 7177 |
+
"entropy": 1.0055284202098846,
|
| 7178 |
+
"epoch": 1.903501029714622,
|
| 7179 |
+
"grad_norm": 0.6064245104789734,
|
| 7180 |
+
"learning_rate": 6.460716536305371e-05,
|
| 7181 |
+
"loss": 1.0159719467163086,
|
| 7182 |
+
"mean_token_accuracy": 0.7620534121990203,
|
| 7183 |
+
"num_tokens": 26128624.0,
|
| 7184 |
+
"step": 6470
|
| 7185 |
+
},
|
| 7186 |
+
{
|
| 7187 |
+
"entropy": 0.9730878472328186,
|
| 7188 |
+
"epoch": 1.9064430714916152,
|
| 7189 |
+
"grad_norm": 0.5835123062133789,
|
| 7190 |
+
"learning_rate": 6.430403362009148e-05,
|
| 7191 |
+
"loss": 0.9895232200622559,
|
| 7192 |
+
"mean_token_accuracy": 0.7685076415538787,
|
| 7193 |
+
"num_tokens": 26168887.0,
|
| 7194 |
+
"step": 6480
|
| 7195 |
+
},
|
| 7196 |
+
{
|
| 7197 |
+
"entropy": 1.0025964260101319,
|
| 7198 |
+
"epoch": 1.9093851132686084,
|
| 7199 |
+
"grad_norm": 0.5540186762809753,
|
| 7200 |
+
"learning_rate": 6.400127731658288e-05,
|
| 7201 |
+
"loss": 1.0111456871032716,
|
| 7202 |
+
"mean_token_accuracy": 0.7678469300270081,
|
| 7203 |
+
"num_tokens": 26209115.0,
|
| 7204 |
+
"step": 6490
|
| 7205 |
+
},
|
| 7206 |
+
{
|
| 7207 |
+
"entropy": 0.9000465154647828,
|
| 7208 |
+
"epoch": 1.9123271550456016,
|
| 7209 |
+
"grad_norm": 0.6538901329040527,
|
| 7210 |
+
"learning_rate": 6.369889963682775e-05,
|
| 7211 |
+
"loss": 0.8854584693908691,
|
| 7212 |
+
"mean_token_accuracy": 0.7922401070594788,
|
| 7213 |
+
"num_tokens": 26249571.0,
|
| 7214 |
+
"step": 6500
|
| 7215 |
+
},
|
| 7216 |
+
{
|
| 7217 |
+
"epoch": 1.9123271550456016,
|
| 7218 |
+
"eval_entropy": 1.0023321302856987,
|
| 7219 |
+
"eval_loss": 1.144269347190857,
|
| 7220 |
+
"eval_mean_token_accuracy": 0.750116533494684,
|
| 7221 |
+
"eval_num_tokens": 26249571.0,
|
| 7222 |
+
"eval_runtime": 116.9879,
|
| 7223 |
+
"eval_samples_per_second": 26.028,
|
| 7224 |
+
"eval_steps_per_second": 3.257,
|
| 7225 |
+
"step": 6500
|
| 7226 |
+
},
|
| 7227 |
+
{
|
| 7228 |
+
"entropy": 0.8953414797782898,
|
| 7229 |
+
"epoch": 1.915269196822595,
|
| 7230 |
+
"grad_norm": 0.653469979763031,
|
| 7231 |
+
"learning_rate": 6.339690376114376e-05,
|
| 7232 |
+
"loss": 0.9109439849853516,
|
| 7233 |
+
"mean_token_accuracy": 0.7873705565929413,
|
| 7234 |
+
"num_tokens": 26289741.0,
|
| 7235 |
+
"step": 6510
|
| 7236 |
+
},
|
| 7237 |
+
{
|
| 7238 |
+
"entropy": 0.9372224569320678,
|
| 7239 |
+
"epoch": 1.918211238599588,
|
| 7240 |
+
"grad_norm": 0.5968233942985535,
|
| 7241 |
+
"learning_rate": 6.309529286583277e-05,
|
| 7242 |
+
"loss": 0.9563371658325195,
|
| 7243 |
+
"mean_token_accuracy": 0.7744209468364716,
|
| 7244 |
+
"num_tokens": 26330177.0,
|
| 7245 |
+
"step": 6520
|
| 7246 |
+
},
|
| 7247 |
+
{
|
| 7248 |
+
"entropy": 0.9607797980308532,
|
| 7249 |
+
"epoch": 1.9211532803765814,
|
| 7250 |
+
"grad_norm": 0.5552541613578796,
|
| 7251 |
+
"learning_rate": 6.279407012314767e-05,
|
| 7252 |
+
"loss": 0.9554115295410156,
|
| 7253 |
+
"mean_token_accuracy": 0.7792925059795379,
|
| 7254 |
+
"num_tokens": 26370862.0,
|
| 7255 |
+
"step": 6530
|
| 7256 |
+
},
|
| 7257 |
+
{
|
| 7258 |
+
"entropy": 0.907786774635315,
|
| 7259 |
+
"epoch": 1.9240953221535746,
|
| 7260 |
+
"grad_norm": 0.5680334568023682,
|
| 7261 |
+
"learning_rate": 6.24932387012587e-05,
|
| 7262 |
+
"loss": 0.9195439338684082,
|
| 7263 |
+
"mean_token_accuracy": 0.7899148881435394,
|
| 7264 |
+
"num_tokens": 26411179.0,
|
| 7265 |
+
"step": 6540
|
| 7266 |
+
},
|
| 7267 |
+
{
|
| 7268 |
+
"entropy": 0.9673022449016571,
|
| 7269 |
+
"epoch": 1.9270373639305678,
|
| 7270 |
+
"grad_norm": 0.6412580013275146,
|
| 7271 |
+
"learning_rate": 6.219280176422049e-05,
|
| 7272 |
+
"loss": 0.954334831237793,
|
| 7273 |
+
"mean_token_accuracy": 0.7745688319206238,
|
| 7274 |
+
"num_tokens": 26451456.0,
|
| 7275 |
+
"step": 6550
|
| 7276 |
+
},
|
| 7277 |
+
{
|
| 7278 |
+
"entropy": 0.87952721118927,
|
| 7279 |
+
"epoch": 1.9299794057075612,
|
| 7280 |
+
"grad_norm": 0.5724232792854309,
|
| 7281 |
+
"learning_rate": 6.189276247193843e-05,
|
| 7282 |
+
"loss": 0.8992857933044434,
|
| 7283 |
+
"mean_token_accuracy": 0.7907371282577514,
|
| 7284 |
+
"num_tokens": 26491838.0,
|
| 7285 |
+
"step": 6560
|
| 7286 |
+
},
|
| 7287 |
+
{
|
| 7288 |
+
"entropy": 0.9637729346752166,
|
| 7289 |
+
"epoch": 1.9329214474845542,
|
| 7290 |
+
"grad_norm": 0.5687930583953857,
|
| 7291 |
+
"learning_rate": 6.159312398013575e-05,
|
| 7292 |
+
"loss": 0.9747313499450684,
|
| 7293 |
+
"mean_token_accuracy": 0.7751651465892792,
|
| 7294 |
+
"num_tokens": 26532331.0,
|
| 7295 |
+
"step": 6570
|
| 7296 |
+
},
|
| 7297 |
+
{
|
| 7298 |
+
"entropy": 0.8893602311611175,
|
| 7299 |
+
"epoch": 1.9358634892615476,
|
| 7300 |
+
"grad_norm": 0.6103338003158569,
|
| 7301 |
+
"learning_rate": 6.129388944032013e-05,
|
| 7302 |
+
"loss": 0.8748321533203125,
|
| 7303 |
+
"mean_token_accuracy": 0.7961347103118896,
|
| 7304 |
+
"num_tokens": 26572762.0,
|
| 7305 |
+
"step": 6580
|
| 7306 |
+
},
|
| 7307 |
+
{
|
| 7308 |
+
"entropy": 0.9709321618080139,
|
| 7309 |
+
"epoch": 1.9388055310385406,
|
| 7310 |
+
"grad_norm": 0.657652735710144,
|
| 7311 |
+
"learning_rate": 6.0995061999750516e-05,
|
| 7312 |
+
"loss": 0.9933636665344239,
|
| 7313 |
+
"mean_token_accuracy": 0.7684541165828704,
|
| 7314 |
+
"num_tokens": 26613213.0,
|
| 7315 |
+
"step": 6590
|
| 7316 |
+
},
|
| 7317 |
+
{
|
| 7318 |
+
"entropy": 0.8971224725246429,
|
| 7319 |
+
"epoch": 1.941747572815534,
|
| 7320 |
+
"grad_norm": 0.5717299580574036,
|
| 7321 |
+
"learning_rate": 6.069664480140424e-05,
|
| 7322 |
+
"loss": 0.9132408142089844,
|
| 7323 |
+
"mean_token_accuracy": 0.7878471255302429,
|
| 7324 |
+
"num_tokens": 26653598.0,
|
| 7325 |
+
"step": 6600
|
| 7326 |
+
},
|
| 7327 |
+
{
|
| 7328 |
+
"epoch": 1.941747572815534,
|
| 7329 |
+
"eval_entropy": 0.9943520688009387,
|
| 7330 |
+
"eval_loss": 1.1435333490371704,
|
| 7331 |
+
"eval_mean_token_accuracy": 0.7505803089442216,
|
| 7332 |
+
"eval_num_tokens": 26653598.0,
|
| 7333 |
+
"eval_runtime": 117.1237,
|
| 7334 |
+
"eval_samples_per_second": 25.998,
|
| 7335 |
+
"eval_steps_per_second": 3.253,
|
| 7336 |
+
"step": 6600
|
| 7337 |
+
},
|
| 7338 |
+
{
|
| 7339 |
+
"entropy": 0.9399532020092011,
|
| 7340 |
+
"epoch": 1.9446896145925272,
|
| 7341 |
+
"grad_norm": 0.6174861788749695,
|
| 7342 |
+
"learning_rate": 6.0398640983943745e-05,
|
| 7343 |
+
"loss": 0.9668970108032227,
|
| 7344 |
+
"mean_token_accuracy": 0.7775733530521393,
|
| 7345 |
+
"num_tokens": 26693838.0,
|
| 7346 |
+
"step": 6610
|
| 7347 |
+
},
|
| 7348 |
+
{
|
| 7349 |
+
"entropy": 1.0020717442035676,
|
| 7350 |
+
"epoch": 1.9476316563695204,
|
| 7351 |
+
"grad_norm": 0.5993858575820923,
|
| 7352 |
+
"learning_rate": 6.0101053681683684e-05,
|
| 7353 |
+
"loss": 0.9971044540405274,
|
| 7354 |
+
"mean_token_accuracy": 0.7669116199016571,
|
| 7355 |
+
"num_tokens": 26734003.0,
|
| 7356 |
+
"step": 6620
|
| 7357 |
+
},
|
| 7358 |
+
{
|
| 7359 |
+
"entropy": 1.0606256008148194,
|
| 7360 |
+
"epoch": 1.9505736981465138,
|
| 7361 |
+
"grad_norm": 0.6137609481811523,
|
| 7362 |
+
"learning_rate": 5.980388602455791e-05,
|
| 7363 |
+
"loss": 1.0581299781799316,
|
| 7364 |
+
"mean_token_accuracy": 0.752843850851059,
|
| 7365 |
+
"num_tokens": 26774676.0,
|
| 7366 |
+
"step": 6630
|
| 7367 |
+
},
|
| 7368 |
+
{
|
| 7369 |
+
"entropy": 0.9770977616310119,
|
| 7370 |
+
"epoch": 1.9535157399235068,
|
| 7371 |
+
"grad_norm": 0.5965011715888977,
|
| 7372 |
+
"learning_rate": 5.950714113808663e-05,
|
| 7373 |
+
"loss": 0.9810546875,
|
| 7374 |
+
"mean_token_accuracy": 0.7688040852546691,
|
| 7375 |
+
"num_tokens": 26815034.0,
|
| 7376 |
+
"step": 6640
|
| 7377 |
+
},
|
| 7378 |
+
{
|
| 7379 |
+
"entropy": 0.9412834465503692,
|
| 7380 |
+
"epoch": 1.9564577817005002,
|
| 7381 |
+
"grad_norm": 0.5246943235397339,
|
| 7382 |
+
"learning_rate": 5.921082214334339e-05,
|
| 7383 |
+
"loss": 0.9711008071899414,
|
| 7384 |
+
"mean_token_accuracy": 0.7793717920780182,
|
| 7385 |
+
"num_tokens": 26855573.0,
|
| 7386 |
+
"step": 6650
|
| 7387 |
+
},
|
| 7388 |
+
{
|
| 7389 |
+
"entropy": 0.9855255424976349,
|
| 7390 |
+
"epoch": 1.9593998234774934,
|
| 7391 |
+
"grad_norm": 0.550930917263031,
|
| 7392 |
+
"learning_rate": 5.891493215692243e-05,
|
| 7393 |
+
"loss": 0.9931858062744141,
|
| 7394 |
+
"mean_token_accuracy": 0.7706417441368103,
|
| 7395 |
+
"num_tokens": 26895609.0,
|
| 7396 |
+
"step": 6660
|
| 7397 |
+
},
|
| 7398 |
+
{
|
| 7399 |
+
"entropy": 0.9438577890396118,
|
| 7400 |
+
"epoch": 1.9623418652544866,
|
| 7401 |
+
"grad_norm": 0.5098885297775269,
|
| 7402 |
+
"learning_rate": 5.861947429090572e-05,
|
| 7403 |
+
"loss": 0.9524516105651856,
|
| 7404 |
+
"mean_token_accuracy": 0.7804840981960297,
|
| 7405 |
+
"num_tokens": 26936122.0,
|
| 7406 |
+
"step": 6670
|
| 7407 |
+
},
|
| 7408 |
+
{
|
| 7409 |
+
"entropy": 0.9131796836853028,
|
| 7410 |
+
"epoch": 1.9652839070314798,
|
| 7411 |
+
"grad_norm": 0.5577981472015381,
|
| 7412 |
+
"learning_rate": 5.832445165283038e-05,
|
| 7413 |
+
"loss": 0.9228861808776856,
|
| 7414 |
+
"mean_token_accuracy": 0.78643758893013,
|
| 7415 |
+
"num_tokens": 26976422.0,
|
| 7416 |
+
"step": 6680
|
| 7417 |
+
},
|
| 7418 |
+
{
|
| 7419 |
+
"entropy": 0.9258975267410279,
|
| 7420 |
+
"epoch": 1.968225948808473,
|
| 7421 |
+
"grad_norm": 0.44099509716033936,
|
| 7422 |
+
"learning_rate": 5.8029867345655885e-05,
|
| 7423 |
+
"loss": 0.9167799949645996,
|
| 7424 |
+
"mean_token_accuracy": 0.785828173160553,
|
| 7425 |
+
"num_tokens": 27016819.0,
|
| 7426 |
+
"step": 6690
|
| 7427 |
+
},
|
| 7428 |
+
{
|
| 7429 |
+
"entropy": 0.9321108460426331,
|
| 7430 |
+
"epoch": 1.9711679905854664,
|
| 7431 |
+
"grad_norm": 0.6357897520065308,
|
| 7432 |
+
"learning_rate": 5.773572446773157e-05,
|
| 7433 |
+
"loss": 0.9454483032226563,
|
| 7434 |
+
"mean_token_accuracy": 0.7806357145309448,
|
| 7435 |
+
"num_tokens": 27057488.0,
|
| 7436 |
+
"step": 6700
|
| 7437 |
+
},
|
| 7438 |
+
{
|
| 7439 |
+
"epoch": 1.9711679905854664,
|
| 7440 |
+
"eval_entropy": 0.9999957291942256,
|
| 7441 |
+
"eval_loss": 1.140816330909729,
|
| 7442 |
+
"eval_mean_token_accuracy": 0.7509619468466191,
|
| 7443 |
+
"eval_num_tokens": 27057488.0,
|
| 7444 |
+
"eval_runtime": 116.9853,
|
| 7445 |
+
"eval_samples_per_second": 26.029,
|
| 7446 |
+
"eval_steps_per_second": 3.257,
|
| 7447 |
+
"step": 6700
|
| 7448 |
}
|
| 7449 |
],
|
| 7450 |
"logging_steps": 10,
|
|
|
|
| 7464 |
"attributes": {}
|
| 7465 |
}
|
| 7466 |
},
|
| 7467 |
+
"total_flos": 1.2570357073262346e+18,
|
| 7468 |
"train_batch_size": 8,
|
| 7469 |
"trial_name": null,
|
| 7470 |
"trial_params": null
|