Training in progress, step 7678, 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 14764744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a8e0149dbe4a109ad1183fd6d8ede3ca18033ce0b6a259c68d2474756655c54
|
| 3 |
size 14764744
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 29614667
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a59a45a19b2c9f23e56e55a71519e761f8f97d3f02fdd625a1d51f610875e79e
|
| 3 |
size 29614667
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3cb25d6d7f0dbf736fabb92aa824ebea295896324cce0f06d6abd8cf16db4d21
|
| 3 |
size 14917
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12f7b6d8d7e8fee907c40bec8ca9a1aa8476f701d70fddd9afd0ce41cbd4305b
|
| 3 |
size 14917
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c6a8c6434d153303d4bf85623f81291d2761c787a467b330cadbbd33803fc16
|
| 3 |
size 1383
|
last-checkpoint/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 sha256:7dbf082c3835739a52a86fc571ce7d0dc13bce7fce7c75fcf407b9af843a96de
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 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,
|
|
@@ -7334,6 +7334,1197 @@
|
|
| 7334 |
"eval_samples_per_second": 5.437,
|
| 7335 |
"eval_steps_per_second": 1.36,
|
| 7336 |
"step": 6600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7337 |
}
|
| 7338 |
],
|
| 7339 |
"logging_steps": 10,
|
|
@@ -7348,12 +8539,12 @@
|
|
| 7348 |
"should_evaluate": false,
|
| 7349 |
"should_log": false,
|
| 7350 |
"should_save": true,
|
| 7351 |
-
"should_training_stop":
|
| 7352 |
},
|
| 7353 |
"attributes": {}
|
| 7354 |
}
|
| 7355 |
},
|
| 7356 |
-
"total_flos": 1.
|
| 7357 |
"train_batch_size": 2,
|
| 7358 |
"trial_name": null,
|
| 7359 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 7678,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 7334 |
"eval_samples_per_second": 5.437,
|
| 7335 |
"eval_steps_per_second": 1.36,
|
| 7336 |
"step": 6600
|
| 7337 |
+
},
|
| 7338 |
+
{
|
| 7339 |
+
"entropy": 0.9823138952255249,
|
| 7340 |
+
"epoch": 1.7218495604037773,
|
| 7341 |
+
"grad_norm": 0.4478851556777954,
|
| 7342 |
+
"learning_rate": 2.7845793175306075e-06,
|
| 7343 |
+
"loss": 0.984638786315918,
|
| 7344 |
+
"mean_token_accuracy": 0.7803937673568726,
|
| 7345 |
+
"num_tokens": 123654.0,
|
| 7346 |
+
"step": 6610
|
| 7347 |
+
},
|
| 7348 |
+
{
|
| 7349 |
+
"entropy": 0.9903601288795472,
|
| 7350 |
+
"epoch": 1.7244545750569849,
|
| 7351 |
+
"grad_norm": 0.4740400016307831,
|
| 7352 |
+
"learning_rate": 2.758530867413389e-06,
|
| 7353 |
+
"loss": 0.9925348281860351,
|
| 7354 |
+
"mean_token_accuracy": 0.7811915218830109,
|
| 7355 |
+
"num_tokens": 250100.0,
|
| 7356 |
+
"step": 6620
|
| 7357 |
+
},
|
| 7358 |
+
{
|
| 7359 |
+
"entropy": 1.0119853645563126,
|
| 7360 |
+
"epoch": 1.7270595897101921,
|
| 7361 |
+
"grad_norm": 0.42794233560562134,
|
| 7362 |
+
"learning_rate": 2.732482417296171e-06,
|
| 7363 |
+
"loss": 1.0036964416503906,
|
| 7364 |
+
"mean_token_accuracy": 0.7763001263141632,
|
| 7365 |
+
"num_tokens": 378633.0,
|
| 7366 |
+
"step": 6630
|
| 7367 |
+
},
|
| 7368 |
+
{
|
| 7369 |
+
"entropy": 0.9493696331977844,
|
| 7370 |
+
"epoch": 1.7296646043633994,
|
| 7371 |
+
"grad_norm": 0.43687593936920166,
|
| 7372 |
+
"learning_rate": 2.706433967178953e-06,
|
| 7373 |
+
"loss": 0.9465467453002929,
|
| 7374 |
+
"mean_token_accuracy": 0.7897346824407577,
|
| 7375 |
+
"num_tokens": 507634.0,
|
| 7376 |
+
"step": 6640
|
| 7377 |
+
},
|
| 7378 |
+
{
|
| 7379 |
+
"entropy": 1.004687362909317,
|
| 7380 |
+
"epoch": 1.732269619016607,
|
| 7381 |
+
"grad_norm": 0.4463178217411041,
|
| 7382 |
+
"learning_rate": 2.680385517061735e-06,
|
| 7383 |
+
"loss": 1.0055112838745117,
|
| 7384 |
+
"mean_token_accuracy": 0.7779219567775726,
|
| 7385 |
+
"num_tokens": 636945.0,
|
| 7386 |
+
"step": 6650
|
| 7387 |
+
},
|
| 7388 |
+
{
|
| 7389 |
+
"entropy": 0.9806973725557327,
|
| 7390 |
+
"epoch": 1.7348746336698144,
|
| 7391 |
+
"grad_norm": 0.4254460334777832,
|
| 7392 |
+
"learning_rate": 2.654337066944517e-06,
|
| 7393 |
+
"loss": 0.9747040748596192,
|
| 7394 |
+
"mean_token_accuracy": 0.7817697480320931,
|
| 7395 |
+
"num_tokens": 766725.0,
|
| 7396 |
+
"step": 6660
|
| 7397 |
+
},
|
| 7398 |
+
{
|
| 7399 |
+
"entropy": 0.9973525032401085,
|
| 7400 |
+
"epoch": 1.737479648323022,
|
| 7401 |
+
"grad_norm": 0.4786764085292816,
|
| 7402 |
+
"learning_rate": 2.6282886168272993e-06,
|
| 7403 |
+
"loss": 1.0005884170532227,
|
| 7404 |
+
"mean_token_accuracy": 0.7777441203594208,
|
| 7405 |
+
"num_tokens": 895306.0,
|
| 7406 |
+
"step": 6670
|
| 7407 |
+
},
|
| 7408 |
+
{
|
| 7409 |
+
"entropy": 0.9907258942723274,
|
| 7410 |
+
"epoch": 1.7400846629762292,
|
| 7411 |
+
"grad_norm": 0.4461391568183899,
|
| 7412 |
+
"learning_rate": 2.602240166710081e-06,
|
| 7413 |
+
"loss": 0.9905254364013671,
|
| 7414 |
+
"mean_token_accuracy": 0.7782276287674904,
|
| 7415 |
+
"num_tokens": 1018780.0,
|
| 7416 |
+
"step": 6680
|
| 7417 |
+
},
|
| 7418 |
+
{
|
| 7419 |
+
"entropy": 0.9964932486414909,
|
| 7420 |
+
"epoch": 1.7426896776294367,
|
| 7421 |
+
"grad_norm": 0.4375130534172058,
|
| 7422 |
+
"learning_rate": 2.576191716592863e-06,
|
| 7423 |
+
"loss": 0.9936812400817872,
|
| 7424 |
+
"mean_token_accuracy": 0.7800995618104934,
|
| 7425 |
+
"num_tokens": 1146973.0,
|
| 7426 |
+
"step": 6690
|
| 7427 |
+
},
|
| 7428 |
+
{
|
| 7429 |
+
"entropy": 0.9998546317219734,
|
| 7430 |
+
"epoch": 1.745294692282644,
|
| 7431 |
+
"grad_norm": 0.5077676177024841,
|
| 7432 |
+
"learning_rate": 2.550143266475645e-06,
|
| 7433 |
+
"loss": 0.9924486160278321,
|
| 7434 |
+
"mean_token_accuracy": 0.7787781774997711,
|
| 7435 |
+
"num_tokens": 1270139.0,
|
| 7436 |
+
"step": 6700
|
| 7437 |
+
},
|
| 7438 |
+
{
|
| 7439 |
+
"epoch": 1.745294692282644,
|
| 7440 |
+
"eval_entropy": 0.988303787839959,
|
| 7441 |
+
"eval_loss": 0.9847902655601501,
|
| 7442 |
+
"eval_mean_token_accuracy": 0.7813749311253271,
|
| 7443 |
+
"eval_num_tokens": 1270139.0,
|
| 7444 |
+
"eval_runtime": 1469.0929,
|
| 7445 |
+
"eval_samples_per_second": 4.633,
|
| 7446 |
+
"eval_steps_per_second": 1.159,
|
| 7447 |
+
"step": 6700
|
| 7448 |
+
},
|
| 7449 |
+
{
|
| 7450 |
+
"entropy": 0.9644292518496513,
|
| 7451 |
+
"epoch": 1.7478997069358515,
|
| 7452 |
+
"grad_norm": 0.49433794617652893,
|
| 7453 |
+
"learning_rate": 2.5240948163584266e-06,
|
| 7454 |
+
"loss": 0.95313720703125,
|
| 7455 |
+
"mean_token_accuracy": 0.7866672441363335,
|
| 7456 |
+
"num_tokens": 1395074.0,
|
| 7457 |
+
"step": 6710
|
| 7458 |
+
},
|
| 7459 |
+
{
|
| 7460 |
+
"entropy": 1.0015375733375549,
|
| 7461 |
+
"epoch": 1.750504721589059,
|
| 7462 |
+
"grad_norm": 0.5286617279052734,
|
| 7463 |
+
"learning_rate": 2.4980463662412086e-06,
|
| 7464 |
+
"loss": 1.0045991897583009,
|
| 7465 |
+
"mean_token_accuracy": 0.7801112249493599,
|
| 7466 |
+
"num_tokens": 1523162.0,
|
| 7467 |
+
"step": 6720
|
| 7468 |
+
},
|
| 7469 |
+
{
|
| 7470 |
+
"entropy": 0.9964019820094109,
|
| 7471 |
+
"epoch": 1.7531097362422665,
|
| 7472 |
+
"grad_norm": 0.5512493252754211,
|
| 7473 |
+
"learning_rate": 2.4719979161239907e-06,
|
| 7474 |
+
"loss": 0.9960369110107422,
|
| 7475 |
+
"mean_token_accuracy": 0.7792898178100586,
|
| 7476 |
+
"num_tokens": 1652179.0,
|
| 7477 |
+
"step": 6730
|
| 7478 |
+
},
|
| 7479 |
+
{
|
| 7480 |
+
"entropy": 0.9651120185852051,
|
| 7481 |
+
"epoch": 1.7557147508954738,
|
| 7482 |
+
"grad_norm": 0.44466137886047363,
|
| 7483 |
+
"learning_rate": 2.4459494660067727e-06,
|
| 7484 |
+
"loss": 0.9617798805236817,
|
| 7485 |
+
"mean_token_accuracy": 0.7847071185708046,
|
| 7486 |
+
"num_tokens": 1780605.0,
|
| 7487 |
+
"step": 6740
|
| 7488 |
+
},
|
| 7489 |
+
{
|
| 7490 |
+
"entropy": 0.9713175401091576,
|
| 7491 |
+
"epoch": 1.758319765548681,
|
| 7492 |
+
"grad_norm": 0.4334625005722046,
|
| 7493 |
+
"learning_rate": 2.4199010158895547e-06,
|
| 7494 |
+
"loss": 0.9693988800048828,
|
| 7495 |
+
"mean_token_accuracy": 0.7834506288170815,
|
| 7496 |
+
"num_tokens": 1907068.0,
|
| 7497 |
+
"step": 6750
|
| 7498 |
+
},
|
| 7499 |
+
{
|
| 7500 |
+
"entropy": 0.9767915159463882,
|
| 7501 |
+
"epoch": 1.7609247802018886,
|
| 7502 |
+
"grad_norm": 0.46460774540901184,
|
| 7503 |
+
"learning_rate": 2.3938525657723368e-06,
|
| 7504 |
+
"loss": 0.9713596343994141,
|
| 7505 |
+
"mean_token_accuracy": 0.783121258020401,
|
| 7506 |
+
"num_tokens": 2036419.0,
|
| 7507 |
+
"step": 6760
|
| 7508 |
+
},
|
| 7509 |
+
{
|
| 7510 |
+
"entropy": 0.9796566858887672,
|
| 7511 |
+
"epoch": 1.763529794855096,
|
| 7512 |
+
"grad_norm": 0.44517257809638977,
|
| 7513 |
+
"learning_rate": 2.367804115655119e-06,
|
| 7514 |
+
"loss": 0.9726932525634766,
|
| 7515 |
+
"mean_token_accuracy": 0.7838401272892952,
|
| 7516 |
+
"num_tokens": 2162634.0,
|
| 7517 |
+
"step": 6770
|
| 7518 |
+
},
|
| 7519 |
+
{
|
| 7520 |
+
"entropy": 1.0208434730768203,
|
| 7521 |
+
"epoch": 1.7661348095083036,
|
| 7522 |
+
"grad_norm": 0.45565611124038696,
|
| 7523 |
+
"learning_rate": 2.341755665537901e-06,
|
| 7524 |
+
"loss": 1.0165723800659179,
|
| 7525 |
+
"mean_token_accuracy": 0.7751889601349831,
|
| 7526 |
+
"num_tokens": 2288676.0,
|
| 7527 |
+
"step": 6780
|
| 7528 |
+
},
|
| 7529 |
+
{
|
| 7530 |
+
"entropy": 0.9510354325175285,
|
| 7531 |
+
"epoch": 1.7687398241615109,
|
| 7532 |
+
"grad_norm": 0.4426063895225525,
|
| 7533 |
+
"learning_rate": 2.315707215420683e-06,
|
| 7534 |
+
"loss": 0.9552410125732422,
|
| 7535 |
+
"mean_token_accuracy": 0.7878434926271438,
|
| 7536 |
+
"num_tokens": 2417023.0,
|
| 7537 |
+
"step": 6790
|
| 7538 |
+
},
|
| 7539 |
+
{
|
| 7540 |
+
"entropy": 1.023421248793602,
|
| 7541 |
+
"epoch": 1.7713448388147184,
|
| 7542 |
+
"grad_norm": 0.45573416352272034,
|
| 7543 |
+
"learning_rate": 2.2896587653034645e-06,
|
| 7544 |
+
"loss": 1.0304821014404297,
|
| 7545 |
+
"mean_token_accuracy": 0.7731378570199012,
|
| 7546 |
+
"num_tokens": 2542335.0,
|
| 7547 |
+
"step": 6800
|
| 7548 |
+
},
|
| 7549 |
+
{
|
| 7550 |
+
"epoch": 1.7713448388147184,
|
| 7551 |
+
"eval_entropy": 0.9875878522875167,
|
| 7552 |
+
"eval_loss": 0.9847276210784912,
|
| 7553 |
+
"eval_mean_token_accuracy": 0.78137964833217,
|
| 7554 |
+
"eval_num_tokens": 2542335.0,
|
| 7555 |
+
"eval_runtime": 1468.1533,
|
| 7556 |
+
"eval_samples_per_second": 4.636,
|
| 7557 |
+
"eval_steps_per_second": 1.159,
|
| 7558 |
+
"step": 6800
|
| 7559 |
+
},
|
| 7560 |
+
{
|
| 7561 |
+
"entropy": 0.9363017573952674,
|
| 7562 |
+
"epoch": 1.7739498534679257,
|
| 7563 |
+
"grad_norm": 0.411294162273407,
|
| 7564 |
+
"learning_rate": 2.2636103151862465e-06,
|
| 7565 |
+
"loss": 0.932148551940918,
|
| 7566 |
+
"mean_token_accuracy": 0.7931686192750931,
|
| 7567 |
+
"num_tokens": 2671515.0,
|
| 7568 |
+
"step": 6810
|
| 7569 |
+
},
|
| 7570 |
+
{
|
| 7571 |
+
"entropy": 1.0269593849778176,
|
| 7572 |
+
"epoch": 1.7765548681211332,
|
| 7573 |
+
"grad_norm": 0.47946611046791077,
|
| 7574 |
+
"learning_rate": 2.2375618650690285e-06,
|
| 7575 |
+
"loss": 1.0208763122558593,
|
| 7576 |
+
"mean_token_accuracy": 0.7766711801290512,
|
| 7577 |
+
"num_tokens": 2796812.0,
|
| 7578 |
+
"step": 6820
|
| 7579 |
+
},
|
| 7580 |
+
{
|
| 7581 |
+
"entropy": 0.9615254417061806,
|
| 7582 |
+
"epoch": 1.7791598827743407,
|
| 7583 |
+
"grad_norm": 0.45359867811203003,
|
| 7584 |
+
"learning_rate": 2.2115134149518106e-06,
|
| 7585 |
+
"loss": 0.9601846694946289,
|
| 7586 |
+
"mean_token_accuracy": 0.7836741924285888,
|
| 7587 |
+
"num_tokens": 2924216.0,
|
| 7588 |
+
"step": 6830
|
| 7589 |
+
},
|
| 7590 |
+
{
|
| 7591 |
+
"entropy": 1.0381957679986953,
|
| 7592 |
+
"epoch": 1.7817648974275482,
|
| 7593 |
+
"grad_norm": 0.4892065227031708,
|
| 7594 |
+
"learning_rate": 2.1854649648345926e-06,
|
| 7595 |
+
"loss": 1.029976749420166,
|
| 7596 |
+
"mean_token_accuracy": 0.7689620569348335,
|
| 7597 |
+
"num_tokens": 3050755.0,
|
| 7598 |
+
"step": 6840
|
| 7599 |
+
},
|
| 7600 |
+
{
|
| 7601 |
+
"entropy": 1.0056249111890794,
|
| 7602 |
+
"epoch": 1.7843699120807555,
|
| 7603 |
+
"grad_norm": 0.5043020844459534,
|
| 7604 |
+
"learning_rate": 2.1594165147173747e-06,
|
| 7605 |
+
"loss": 1.0031335830688477,
|
| 7606 |
+
"mean_token_accuracy": 0.778705857694149,
|
| 7607 |
+
"num_tokens": 3177200.0,
|
| 7608 |
+
"step": 6850
|
| 7609 |
+
},
|
| 7610 |
+
{
|
| 7611 |
+
"entropy": 0.9671053498983383,
|
| 7612 |
+
"epoch": 1.7869749267339627,
|
| 7613 |
+
"grad_norm": 0.4945420026779175,
|
| 7614 |
+
"learning_rate": 2.1333680646001563e-06,
|
| 7615 |
+
"loss": 0.967266845703125,
|
| 7616 |
+
"mean_token_accuracy": 0.7848890349268913,
|
| 7617 |
+
"num_tokens": 3303486.0,
|
| 7618 |
+
"step": 6860
|
| 7619 |
+
},
|
| 7620 |
+
{
|
| 7621 |
+
"entropy": 0.9680422604084015,
|
| 7622 |
+
"epoch": 1.7895799413871702,
|
| 7623 |
+
"grad_norm": 0.4735678732395172,
|
| 7624 |
+
"learning_rate": 2.1073196144829383e-06,
|
| 7625 |
+
"loss": 0.9629220962524414,
|
| 7626 |
+
"mean_token_accuracy": 0.7858682468533515,
|
| 7627 |
+
"num_tokens": 3427886.0,
|
| 7628 |
+
"step": 6870
|
| 7629 |
+
},
|
| 7630 |
+
{
|
| 7631 |
+
"entropy": 0.9950935959815979,
|
| 7632 |
+
"epoch": 1.7921849560403778,
|
| 7633 |
+
"grad_norm": 0.4246147871017456,
|
| 7634 |
+
"learning_rate": 2.0812711643657203e-06,
|
| 7635 |
+
"loss": 0.9934613227844238,
|
| 7636 |
+
"mean_token_accuracy": 0.7778671503067016,
|
| 7637 |
+
"num_tokens": 3558964.0,
|
| 7638 |
+
"step": 6880
|
| 7639 |
+
},
|
| 7640 |
+
{
|
| 7641 |
+
"entropy": 0.9615059599280358,
|
| 7642 |
+
"epoch": 1.7947899706935853,
|
| 7643 |
+
"grad_norm": 0.44503042101860046,
|
| 7644 |
+
"learning_rate": 2.0552227142485024e-06,
|
| 7645 |
+
"loss": 0.9620532989501953,
|
| 7646 |
+
"mean_token_accuracy": 0.7851508602499961,
|
| 7647 |
+
"num_tokens": 3691908.0,
|
| 7648 |
+
"step": 6890
|
| 7649 |
+
},
|
| 7650 |
+
{
|
| 7651 |
+
"entropy": 0.9874796941876411,
|
| 7652 |
+
"epoch": 1.7973949853467925,
|
| 7653 |
+
"grad_norm": 0.4577827751636505,
|
| 7654 |
+
"learning_rate": 2.0291742641312844e-06,
|
| 7655 |
+
"loss": 0.9817806243896484,
|
| 7656 |
+
"mean_token_accuracy": 0.7804968252778053,
|
| 7657 |
+
"num_tokens": 3816343.0,
|
| 7658 |
+
"step": 6900
|
| 7659 |
+
},
|
| 7660 |
+
{
|
| 7661 |
+
"epoch": 1.7973949853467925,
|
| 7662 |
+
"eval_entropy": 0.987803460007969,
|
| 7663 |
+
"eval_loss": 0.9846410155296326,
|
| 7664 |
+
"eval_mean_token_accuracy": 0.7813729251593736,
|
| 7665 |
+
"eval_num_tokens": 3816343.0,
|
| 7666 |
+
"eval_runtime": 1469.6562,
|
| 7667 |
+
"eval_samples_per_second": 4.631,
|
| 7668 |
+
"eval_steps_per_second": 1.158,
|
| 7669 |
+
"step": 6900
|
| 7670 |
+
},
|
| 7671 |
+
{
|
| 7672 |
+
"entropy": 0.9574330061674118,
|
| 7673 |
+
"epoch": 1.8,
|
| 7674 |
+
"grad_norm": 0.44457337260246277,
|
| 7675 |
+
"learning_rate": 2.0031258140140664e-06,
|
| 7676 |
+
"loss": 0.9461252212524414,
|
| 7677 |
+
"mean_token_accuracy": 0.788931405544281,
|
| 7678 |
+
"num_tokens": 3947338.0,
|
| 7679 |
+
"step": 6910
|
| 7680 |
+
},
|
| 7681 |
+
{
|
| 7682 |
+
"entropy": 0.9869294106960297,
|
| 7683 |
+
"epoch": 1.8026050146532073,
|
| 7684 |
+
"grad_norm": 0.46561387181282043,
|
| 7685 |
+
"learning_rate": 1.977077363896848e-06,
|
| 7686 |
+
"loss": 0.983763313293457,
|
| 7687 |
+
"mean_token_accuracy": 0.782055202126503,
|
| 7688 |
+
"num_tokens": 4077287.0,
|
| 7689 |
+
"step": 6920
|
| 7690 |
+
},
|
| 7691 |
+
{
|
| 7692 |
+
"entropy": 1.0384597808122635,
|
| 7693 |
+
"epoch": 1.8052100293064148,
|
| 7694 |
+
"grad_norm": 0.4717126488685608,
|
| 7695 |
+
"learning_rate": 1.9510289137796305e-06,
|
| 7696 |
+
"loss": 1.0406241416931152,
|
| 7697 |
+
"mean_token_accuracy": 0.7699192970991134,
|
| 7698 |
+
"num_tokens": 4202626.0,
|
| 7699 |
+
"step": 6930
|
| 7700 |
+
},
|
| 7701 |
+
{
|
| 7702 |
+
"entropy": 0.9926999747753144,
|
| 7703 |
+
"epoch": 1.8078150439596223,
|
| 7704 |
+
"grad_norm": 0.4420601427555084,
|
| 7705 |
+
"learning_rate": 1.924980463662412e-06,
|
| 7706 |
+
"loss": 0.9950467109680176,
|
| 7707 |
+
"mean_token_accuracy": 0.778462678194046,
|
| 7708 |
+
"num_tokens": 4329768.0,
|
| 7709 |
+
"step": 6940
|
| 7710 |
+
},
|
| 7711 |
+
{
|
| 7712 |
+
"entropy": 1.0370307445526123,
|
| 7713 |
+
"epoch": 1.8104200586128298,
|
| 7714 |
+
"grad_norm": 0.44570279121398926,
|
| 7715 |
+
"learning_rate": 1.8989320135451942e-06,
|
| 7716 |
+
"loss": 1.0290287017822266,
|
| 7717 |
+
"mean_token_accuracy": 0.7743745654821396,
|
| 7718 |
+
"num_tokens": 4454595.0,
|
| 7719 |
+
"step": 6950
|
| 7720 |
+
},
|
| 7721 |
+
{
|
| 7722 |
+
"entropy": 0.9640106305480003,
|
| 7723 |
+
"epoch": 1.8130250732660371,
|
| 7724 |
+
"grad_norm": 0.5364393591880798,
|
| 7725 |
+
"learning_rate": 1.8728835634279762e-06,
|
| 7726 |
+
"loss": 0.9536015510559082,
|
| 7727 |
+
"mean_token_accuracy": 0.786224502325058,
|
| 7728 |
+
"num_tokens": 4579180.0,
|
| 7729 |
+
"step": 6960
|
| 7730 |
+
},
|
| 7731 |
+
{
|
| 7732 |
+
"entropy": 1.0095271036028861,
|
| 7733 |
+
"epoch": 1.8156300879192444,
|
| 7734 |
+
"grad_norm": 0.4075166583061218,
|
| 7735 |
+
"learning_rate": 1.8468351133107582e-06,
|
| 7736 |
+
"loss": 0.9975923538208008,
|
| 7737 |
+
"mean_token_accuracy": 0.7763581544160842,
|
| 7738 |
+
"num_tokens": 4707103.0,
|
| 7739 |
+
"step": 6970
|
| 7740 |
+
},
|
| 7741 |
+
{
|
| 7742 |
+
"entropy": 1.0112810552120208,
|
| 7743 |
+
"epoch": 1.818235102572452,
|
| 7744 |
+
"grad_norm": 0.4442594647407532,
|
| 7745 |
+
"learning_rate": 1.82078666319354e-06,
|
| 7746 |
+
"loss": 1.0171551704406738,
|
| 7747 |
+
"mean_token_accuracy": 0.7736199840903282,
|
| 7748 |
+
"num_tokens": 4835159.0,
|
| 7749 |
+
"step": 6980
|
| 7750 |
+
},
|
| 7751 |
+
{
|
| 7752 |
+
"entropy": 0.9952105522155762,
|
| 7753 |
+
"epoch": 1.8208401172256594,
|
| 7754 |
+
"grad_norm": 0.4370501935482025,
|
| 7755 |
+
"learning_rate": 1.794738213076322e-06,
|
| 7756 |
+
"loss": 0.9866777420043945,
|
| 7757 |
+
"mean_token_accuracy": 0.780173048377037,
|
| 7758 |
+
"num_tokens": 4963278.0,
|
| 7759 |
+
"step": 6990
|
| 7760 |
+
},
|
| 7761 |
+
{
|
| 7762 |
+
"entropy": 1.0098994955420495,
|
| 7763 |
+
"epoch": 1.823445131878867,
|
| 7764 |
+
"grad_norm": 0.4337042272090912,
|
| 7765 |
+
"learning_rate": 1.7686897629591041e-06,
|
| 7766 |
+
"loss": 1.0060459136962892,
|
| 7767 |
+
"mean_token_accuracy": 0.7759813785552978,
|
| 7768 |
+
"num_tokens": 5088254.0,
|
| 7769 |
+
"step": 7000
|
| 7770 |
+
},
|
| 7771 |
+
{
|
| 7772 |
+
"epoch": 1.823445131878867,
|
| 7773 |
+
"eval_entropy": 0.9862276635775135,
|
| 7774 |
+
"eval_loss": 0.984578013420105,
|
| 7775 |
+
"eval_mean_token_accuracy": 0.7813913609950719,
|
| 7776 |
+
"eval_num_tokens": 5088254.0,
|
| 7777 |
+
"eval_runtime": 1468.0381,
|
| 7778 |
+
"eval_samples_per_second": 4.636,
|
| 7779 |
+
"eval_steps_per_second": 1.159,
|
| 7780 |
+
"step": 7000
|
| 7781 |
+
},
|
| 7782 |
+
{
|
| 7783 |
+
"entropy": 1.0100476831197738,
|
| 7784 |
+
"epoch": 1.8260501465320742,
|
| 7785 |
+
"grad_norm": 0.44410353899002075,
|
| 7786 |
+
"learning_rate": 1.742641312841886e-06,
|
| 7787 |
+
"loss": 1.000528907775879,
|
| 7788 |
+
"mean_token_accuracy": 0.7768819153308868,
|
| 7789 |
+
"num_tokens": 5217573.0,
|
| 7790 |
+
"step": 7010
|
| 7791 |
+
},
|
| 7792 |
+
{
|
| 7793 |
+
"entropy": 0.9736192256212235,
|
| 7794 |
+
"epoch": 1.8286551611852817,
|
| 7795 |
+
"grad_norm": 0.773119330406189,
|
| 7796 |
+
"learning_rate": 1.716592862724668e-06,
|
| 7797 |
+
"loss": 0.9745948791503907,
|
| 7798 |
+
"mean_token_accuracy": 0.7835270285606384,
|
| 7799 |
+
"num_tokens": 5338904.0,
|
| 7800 |
+
"step": 7020
|
| 7801 |
+
},
|
| 7802 |
+
{
|
| 7803 |
+
"entropy": 0.9792160719633103,
|
| 7804 |
+
"epoch": 1.831260175838489,
|
| 7805 |
+
"grad_norm": 0.4685461223125458,
|
| 7806 |
+
"learning_rate": 1.69054441260745e-06,
|
| 7807 |
+
"loss": 0.9773174285888672,
|
| 7808 |
+
"mean_token_accuracy": 0.7809088915586472,
|
| 7809 |
+
"num_tokens": 5465329.0,
|
| 7810 |
+
"step": 7030
|
| 7811 |
+
},
|
| 7812 |
+
{
|
| 7813 |
+
"entropy": 1.0000706896185876,
|
| 7814 |
+
"epoch": 1.8338651904916965,
|
| 7815 |
+
"grad_norm": 0.5007032155990601,
|
| 7816 |
+
"learning_rate": 1.664495962490232e-06,
|
| 7817 |
+
"loss": 0.9990605354309082,
|
| 7818 |
+
"mean_token_accuracy": 0.7788797214627265,
|
| 7819 |
+
"num_tokens": 5593851.0,
|
| 7820 |
+
"step": 7040
|
| 7821 |
+
},
|
| 7822 |
+
{
|
| 7823 |
+
"entropy": 0.988054946064949,
|
| 7824 |
+
"epoch": 1.836470205144904,
|
| 7825 |
+
"grad_norm": 0.4550478756427765,
|
| 7826 |
+
"learning_rate": 1.6384475123730139e-06,
|
| 7827 |
+
"loss": 0.9810928344726563,
|
| 7828 |
+
"mean_token_accuracy": 0.7814757525920868,
|
| 7829 |
+
"num_tokens": 5723662.0,
|
| 7830 |
+
"step": 7050
|
| 7831 |
+
},
|
| 7832 |
+
{
|
| 7833 |
+
"entropy": 0.9801774248480797,
|
| 7834 |
+
"epoch": 1.8390752197981115,
|
| 7835 |
+
"grad_norm": 0.4820897579193115,
|
| 7836 |
+
"learning_rate": 1.612399062255796e-06,
|
| 7837 |
+
"loss": 0.9793455123901367,
|
| 7838 |
+
"mean_token_accuracy": 0.7816591963171959,
|
| 7839 |
+
"num_tokens": 5849925.0,
|
| 7840 |
+
"step": 7060
|
| 7841 |
+
},
|
| 7842 |
+
{
|
| 7843 |
+
"entropy": 1.023764932155609,
|
| 7844 |
+
"epoch": 1.8416802344513188,
|
| 7845 |
+
"grad_norm": 0.4377996623516083,
|
| 7846 |
+
"learning_rate": 1.586350612138578e-06,
|
| 7847 |
+
"loss": 1.0208126068115235,
|
| 7848 |
+
"mean_token_accuracy": 0.772579737007618,
|
| 7849 |
+
"num_tokens": 5977074.0,
|
| 7850 |
+
"step": 7070
|
| 7851 |
+
},
|
| 7852 |
+
{
|
| 7853 |
+
"entropy": 0.9775882303714752,
|
| 7854 |
+
"epoch": 1.844285249104526,
|
| 7855 |
+
"grad_norm": 0.5249976515769958,
|
| 7856 |
+
"learning_rate": 1.56030216202136e-06,
|
| 7857 |
+
"loss": 0.9740531921386719,
|
| 7858 |
+
"mean_token_accuracy": 0.7832024395465851,
|
| 7859 |
+
"num_tokens": 6105808.0,
|
| 7860 |
+
"step": 7080
|
| 7861 |
+
},
|
| 7862 |
+
{
|
| 7863 |
+
"entropy": 1.023333878815174,
|
| 7864 |
+
"epoch": 1.8468902637577336,
|
| 7865 |
+
"grad_norm": 0.45492520928382874,
|
| 7866 |
+
"learning_rate": 1.5342537119041418e-06,
|
| 7867 |
+
"loss": 1.0202534675598145,
|
| 7868 |
+
"mean_token_accuracy": 0.7749065518379211,
|
| 7869 |
+
"num_tokens": 6232996.0,
|
| 7870 |
+
"step": 7090
|
| 7871 |
+
},
|
| 7872 |
+
{
|
| 7873 |
+
"entropy": 0.9751718774437904,
|
| 7874 |
+
"epoch": 1.849495278410941,
|
| 7875 |
+
"grad_norm": 0.4671773314476013,
|
| 7876 |
+
"learning_rate": 1.5082052617869238e-06,
|
| 7877 |
+
"loss": 0.9715238571166992,
|
| 7878 |
+
"mean_token_accuracy": 0.7816667452454567,
|
| 7879 |
+
"num_tokens": 6358515.0,
|
| 7880 |
+
"step": 7100
|
| 7881 |
+
},
|
| 7882 |
+
{
|
| 7883 |
+
"epoch": 1.849495278410941,
|
| 7884 |
+
"eval_entropy": 0.984380263314264,
|
| 7885 |
+
"eval_loss": 0.9845257997512817,
|
| 7886 |
+
"eval_mean_token_accuracy": 0.781415506409422,
|
| 7887 |
+
"eval_num_tokens": 6358515.0,
|
| 7888 |
+
"eval_runtime": 1468.7893,
|
| 7889 |
+
"eval_samples_per_second": 4.634,
|
| 7890 |
+
"eval_steps_per_second": 1.159,
|
| 7891 |
+
"step": 7100
|
| 7892 |
+
},
|
| 7893 |
+
{
|
| 7894 |
+
"entropy": 0.974171893298626,
|
| 7895 |
+
"epoch": 1.8521002930641486,
|
| 7896 |
+
"grad_norm": 0.4646225571632385,
|
| 7897 |
+
"learning_rate": 1.4821568116697059e-06,
|
| 7898 |
+
"loss": 0.9782926559448242,
|
| 7899 |
+
"mean_token_accuracy": 0.7838637784123421,
|
| 7900 |
+
"num_tokens": 6484420.0,
|
| 7901 |
+
"step": 7110
|
| 7902 |
+
},
|
| 7903 |
+
{
|
| 7904 |
+
"entropy": 0.9854314371943473,
|
| 7905 |
+
"epoch": 1.8547053077173559,
|
| 7906 |
+
"grad_norm": 0.48577308654785156,
|
| 7907 |
+
"learning_rate": 1.4561083615524877e-06,
|
| 7908 |
+
"loss": 0.9864839553833008,
|
| 7909 |
+
"mean_token_accuracy": 0.7816243350505829,
|
| 7910 |
+
"num_tokens": 6613748.0,
|
| 7911 |
+
"step": 7120
|
| 7912 |
+
},
|
| 7913 |
+
{
|
| 7914 |
+
"entropy": 1.0110283181071282,
|
| 7915 |
+
"epoch": 1.8573103223705634,
|
| 7916 |
+
"grad_norm": 0.45445773005485535,
|
| 7917 |
+
"learning_rate": 1.4300599114352697e-06,
|
| 7918 |
+
"loss": 1.0078933715820313,
|
| 7919 |
+
"mean_token_accuracy": 0.7752061516046524,
|
| 7920 |
+
"num_tokens": 6742867.0,
|
| 7921 |
+
"step": 7130
|
| 7922 |
+
},
|
| 7923 |
+
{
|
| 7924 |
+
"entropy": 0.9768914580345154,
|
| 7925 |
+
"epoch": 1.8599153370237707,
|
| 7926 |
+
"grad_norm": 0.49669763445854187,
|
| 7927 |
+
"learning_rate": 1.4040114613180518e-06,
|
| 7928 |
+
"loss": 0.9594968795776367,
|
| 7929 |
+
"mean_token_accuracy": 0.7865890443325043,
|
| 7930 |
+
"num_tokens": 6869550.0,
|
| 7931 |
+
"step": 7140
|
| 7932 |
+
},
|
| 7933 |
+
{
|
| 7934 |
+
"entropy": 0.9895724222064018,
|
| 7935 |
+
"epoch": 1.8625203516769782,
|
| 7936 |
+
"grad_norm": 0.49469998478889465,
|
| 7937 |
+
"learning_rate": 1.3779630112008336e-06,
|
| 7938 |
+
"loss": 0.9920047760009766,
|
| 7939 |
+
"mean_token_accuracy": 0.7797763243317604,
|
| 7940 |
+
"num_tokens": 6993128.0,
|
| 7941 |
+
"step": 7150
|
| 7942 |
+
},
|
| 7943 |
+
{
|
| 7944 |
+
"entropy": 0.986992497742176,
|
| 7945 |
+
"epoch": 1.8651253663301857,
|
| 7946 |
+
"grad_norm": 0.44448405504226685,
|
| 7947 |
+
"learning_rate": 1.3519145610836156e-06,
|
| 7948 |
+
"loss": 0.9805912971496582,
|
| 7949 |
+
"mean_token_accuracy": 0.7820166110992431,
|
| 7950 |
+
"num_tokens": 7121816.0,
|
| 7951 |
+
"step": 7160
|
| 7952 |
+
},
|
| 7953 |
+
{
|
| 7954 |
+
"entropy": 1.0394649550318718,
|
| 7955 |
+
"epoch": 1.8677303809833932,
|
| 7956 |
+
"grad_norm": 0.44717878103256226,
|
| 7957 |
+
"learning_rate": 1.3258661109663976e-06,
|
| 7958 |
+
"loss": 1.0331060409545898,
|
| 7959 |
+
"mean_token_accuracy": 0.7710637152194977,
|
| 7960 |
+
"num_tokens": 7251084.0,
|
| 7961 |
+
"step": 7170
|
| 7962 |
+
},
|
| 7963 |
+
{
|
| 7964 |
+
"entropy": 0.9745459094643593,
|
| 7965 |
+
"epoch": 1.8703353956366005,
|
| 7966 |
+
"grad_norm": 0.45827585458755493,
|
| 7967 |
+
"learning_rate": 1.2998176608491797e-06,
|
| 7968 |
+
"loss": 0.9728391647338868,
|
| 7969 |
+
"mean_token_accuracy": 0.7855743259191513,
|
| 7970 |
+
"num_tokens": 7377259.0,
|
| 7971 |
+
"step": 7180
|
| 7972 |
+
},
|
| 7973 |
+
{
|
| 7974 |
+
"entropy": 0.9786895856261253,
|
| 7975 |
+
"epoch": 1.8729404102898077,
|
| 7976 |
+
"grad_norm": 0.5126107335090637,
|
| 7977 |
+
"learning_rate": 1.2737692107319615e-06,
|
| 7978 |
+
"loss": 0.9728766441345215,
|
| 7979 |
+
"mean_token_accuracy": 0.7842805728316307,
|
| 7980 |
+
"num_tokens": 7505756.0,
|
| 7981 |
+
"step": 7190
|
| 7982 |
+
},
|
| 7983 |
+
{
|
| 7984 |
+
"entropy": 0.9696195796132088,
|
| 7985 |
+
"epoch": 1.8755454249430152,
|
| 7986 |
+
"grad_norm": 0.43363329768180847,
|
| 7987 |
+
"learning_rate": 1.2477207606147435e-06,
|
| 7988 |
+
"loss": 0.9623435974121094,
|
| 7989 |
+
"mean_token_accuracy": 0.7858400091528892,
|
| 7990 |
+
"num_tokens": 7635894.0,
|
| 7991 |
+
"step": 7200
|
| 7992 |
+
},
|
| 7993 |
+
{
|
| 7994 |
+
"epoch": 1.8755454249430152,
|
| 7995 |
+
"eval_entropy": 0.986146639117623,
|
| 7996 |
+
"eval_loss": 0.9844483137130737,
|
| 7997 |
+
"eval_mean_token_accuracy": 0.7814126177904328,
|
| 7998 |
+
"eval_num_tokens": 7635894.0,
|
| 7999 |
+
"eval_runtime": 1467.7106,
|
| 8000 |
+
"eval_samples_per_second": 4.637,
|
| 8001 |
+
"eval_steps_per_second": 1.16,
|
| 8002 |
+
"step": 7200
|
| 8003 |
+
},
|
| 8004 |
+
{
|
| 8005 |
+
"entropy": 1.0052977442741393,
|
| 8006 |
+
"epoch": 1.8781504395962227,
|
| 8007 |
+
"grad_norm": 0.48793092370033264,
|
| 8008 |
+
"learning_rate": 1.2216723104975254e-06,
|
| 8009 |
+
"loss": 0.991578483581543,
|
| 8010 |
+
"mean_token_accuracy": 0.7797325789928437,
|
| 8011 |
+
"num_tokens": 7765956.0,
|
| 8012 |
+
"step": 7210
|
| 8013 |
+
},
|
| 8014 |
+
{
|
| 8015 |
+
"entropy": 0.9644748240709304,
|
| 8016 |
+
"epoch": 1.8807554542494302,
|
| 8017 |
+
"grad_norm": 0.5126164555549622,
|
| 8018 |
+
"learning_rate": 1.1956238603803074e-06,
|
| 8019 |
+
"loss": 0.9693941116333008,
|
| 8020 |
+
"mean_token_accuracy": 0.7834710210561753,
|
| 8021 |
+
"num_tokens": 7894428.0,
|
| 8022 |
+
"step": 7220
|
| 8023 |
+
},
|
| 8024 |
+
{
|
| 8025 |
+
"entropy": 0.9468156561255455,
|
| 8026 |
+
"epoch": 1.8833604689026375,
|
| 8027 |
+
"grad_norm": 0.42623046040534973,
|
| 8028 |
+
"learning_rate": 1.1695754102630894e-06,
|
| 8029 |
+
"loss": 0.9442598342895507,
|
| 8030 |
+
"mean_token_accuracy": 0.790334589779377,
|
| 8031 |
+
"num_tokens": 8025391.0,
|
| 8032 |
+
"step": 7230
|
| 8033 |
+
},
|
| 8034 |
+
{
|
| 8035 |
+
"entropy": 0.9767702788114547,
|
| 8036 |
+
"epoch": 1.885965483555845,
|
| 8037 |
+
"grad_norm": 0.4314723610877991,
|
| 8038 |
+
"learning_rate": 1.1435269601458713e-06,
|
| 8039 |
+
"loss": 0.984138298034668,
|
| 8040 |
+
"mean_token_accuracy": 0.7804219737648964,
|
| 8041 |
+
"num_tokens": 8155196.0,
|
| 8042 |
+
"step": 7240
|
| 8043 |
+
},
|
| 8044 |
+
{
|
| 8045 |
+
"entropy": 1.035601581633091,
|
| 8046 |
+
"epoch": 1.8885704982090523,
|
| 8047 |
+
"grad_norm": 0.48795798420906067,
|
| 8048 |
+
"learning_rate": 1.1174785100286533e-06,
|
| 8049 |
+
"loss": 1.0341554641723634,
|
| 8050 |
+
"mean_token_accuracy": 0.770530553162098,
|
| 8051 |
+
"num_tokens": 8282492.0,
|
| 8052 |
+
"step": 7250
|
| 8053 |
+
},
|
| 8054 |
+
{
|
| 8055 |
+
"entropy": 0.9517062842845917,
|
| 8056 |
+
"epoch": 1.8911755128622598,
|
| 8057 |
+
"grad_norm": 0.5049936771392822,
|
| 8058 |
+
"learning_rate": 1.0914300599114353e-06,
|
| 8059 |
+
"loss": 0.9471035957336426,
|
| 8060 |
+
"mean_token_accuracy": 0.787438040971756,
|
| 8061 |
+
"num_tokens": 8405078.0,
|
| 8062 |
+
"step": 7260
|
| 8063 |
+
},
|
| 8064 |
+
{
|
| 8065 |
+
"entropy": 1.0409354090690612,
|
| 8066 |
+
"epoch": 1.8937805275154673,
|
| 8067 |
+
"grad_norm": 0.41695478558540344,
|
| 8068 |
+
"learning_rate": 1.0653816097942174e-06,
|
| 8069 |
+
"loss": 1.03995361328125,
|
| 8070 |
+
"mean_token_accuracy": 0.7707146346569062,
|
| 8071 |
+
"num_tokens": 8534959.0,
|
| 8072 |
+
"step": 7270
|
| 8073 |
+
},
|
| 8074 |
+
{
|
| 8075 |
+
"entropy": 1.0195857673883437,
|
| 8076 |
+
"epoch": 1.8963855421686748,
|
| 8077 |
+
"grad_norm": 0.4800625443458557,
|
| 8078 |
+
"learning_rate": 1.0393331596769992e-06,
|
| 8079 |
+
"loss": 1.031193161010742,
|
| 8080 |
+
"mean_token_accuracy": 0.775109025835991,
|
| 8081 |
+
"num_tokens": 8663313.0,
|
| 8082 |
+
"step": 7280
|
| 8083 |
+
},
|
| 8084 |
+
{
|
| 8085 |
+
"entropy": 0.9964761912822724,
|
| 8086 |
+
"epoch": 1.8989905568218821,
|
| 8087 |
+
"grad_norm": 0.48612233996391296,
|
| 8088 |
+
"learning_rate": 1.0132847095597812e-06,
|
| 8089 |
+
"loss": 0.9887420654296875,
|
| 8090 |
+
"mean_token_accuracy": 0.780465230345726,
|
| 8091 |
+
"num_tokens": 8789142.0,
|
| 8092 |
+
"step": 7290
|
| 8093 |
+
},
|
| 8094 |
+
{
|
| 8095 |
+
"entropy": 0.990180803835392,
|
| 8096 |
+
"epoch": 1.9015955714750894,
|
| 8097 |
+
"grad_norm": 0.48173290491104126,
|
| 8098 |
+
"learning_rate": 9.872362594425633e-07,
|
| 8099 |
+
"loss": 0.9843120574951172,
|
| 8100 |
+
"mean_token_accuracy": 0.780947034060955,
|
| 8101 |
+
"num_tokens": 8917921.0,
|
| 8102 |
+
"step": 7300
|
| 8103 |
+
},
|
| 8104 |
+
{
|
| 8105 |
+
"epoch": 1.9015955714750894,
|
| 8106 |
+
"eval_entropy": 0.9868813281403865,
|
| 8107 |
+
"eval_loss": 0.9844101667404175,
|
| 8108 |
+
"eval_mean_token_accuracy": 0.7814485650995503,
|
| 8109 |
+
"eval_num_tokens": 8917921.0,
|
| 8110 |
+
"eval_runtime": 1467.1092,
|
| 8111 |
+
"eval_samples_per_second": 4.639,
|
| 8112 |
+
"eval_steps_per_second": 1.16,
|
| 8113 |
+
"step": 7300
|
| 8114 |
+
},
|
| 8115 |
+
{
|
| 8116 |
+
"entropy": 0.9696540087461472,
|
| 8117 |
+
"epoch": 1.904200586128297,
|
| 8118 |
+
"grad_norm": 0.5174248218536377,
|
| 8119 |
+
"learning_rate": 9.61187809325345e-07,
|
| 8120 |
+
"loss": 0.9633716583251953,
|
| 8121 |
+
"mean_token_accuracy": 0.7845815554261207,
|
| 8122 |
+
"num_tokens": 9041505.0,
|
| 8123 |
+
"step": 7310
|
| 8124 |
+
},
|
| 8125 |
+
{
|
| 8126 |
+
"entropy": 1.01645025908947,
|
| 8127 |
+
"epoch": 1.9068056007815044,
|
| 8128 |
+
"grad_norm": 0.5100550651550293,
|
| 8129 |
+
"learning_rate": 9.351393592081272e-07,
|
| 8130 |
+
"loss": 1.0193856239318848,
|
| 8131 |
+
"mean_token_accuracy": 0.7724874794483185,
|
| 8132 |
+
"num_tokens": 9169346.0,
|
| 8133 |
+
"step": 7320
|
| 8134 |
+
},
|
| 8135 |
+
{
|
| 8136 |
+
"entropy": 0.9561196148395539,
|
| 8137 |
+
"epoch": 1.909410615434712,
|
| 8138 |
+
"grad_norm": 0.4802747368812561,
|
| 8139 |
+
"learning_rate": 9.090909090909091e-07,
|
| 8140 |
+
"loss": 0.9532204627990722,
|
| 8141 |
+
"mean_token_accuracy": 0.7851155206561089,
|
| 8142 |
+
"num_tokens": 9295095.0,
|
| 8143 |
+
"step": 7330
|
| 8144 |
+
},
|
| 8145 |
+
{
|
| 8146 |
+
"entropy": 0.9658488750457763,
|
| 8147 |
+
"epoch": 1.9120156300879192,
|
| 8148 |
+
"grad_norm": 0.45807015895843506,
|
| 8149 |
+
"learning_rate": 8.830424589736912e-07,
|
| 8150 |
+
"loss": 0.9590003967285157,
|
| 8151 |
+
"mean_token_accuracy": 0.7856974363327026,
|
| 8152 |
+
"num_tokens": 9422841.0,
|
| 8153 |
+
"step": 7340
|
| 8154 |
+
},
|
| 8155 |
+
{
|
| 8156 |
+
"entropy": 0.9856777518987656,
|
| 8157 |
+
"epoch": 1.9146206447411267,
|
| 8158 |
+
"grad_norm": 0.4914281964302063,
|
| 8159 |
+
"learning_rate": 8.569940088564731e-07,
|
| 8160 |
+
"loss": 0.978643798828125,
|
| 8161 |
+
"mean_token_accuracy": 0.7827108010649682,
|
| 8162 |
+
"num_tokens": 9549389.0,
|
| 8163 |
+
"step": 7350
|
| 8164 |
+
},
|
| 8165 |
+
{
|
| 8166 |
+
"entropy": 0.9775399908423423,
|
| 8167 |
+
"epoch": 1.917225659394334,
|
| 8168 |
+
"grad_norm": 0.44177767634391785,
|
| 8169 |
+
"learning_rate": 8.30945558739255e-07,
|
| 8170 |
+
"loss": 0.9639266967773438,
|
| 8171 |
+
"mean_token_accuracy": 0.7834463134407997,
|
| 8172 |
+
"num_tokens": 9673807.0,
|
| 8173 |
+
"step": 7360
|
| 8174 |
+
},
|
| 8175 |
+
{
|
| 8176 |
+
"entropy": 0.9876919373869896,
|
| 8177 |
+
"epoch": 1.9198306740475415,
|
| 8178 |
+
"grad_norm": 0.45494985580444336,
|
| 8179 |
+
"learning_rate": 8.048971086220371e-07,
|
| 8180 |
+
"loss": 0.9917585372924804,
|
| 8181 |
+
"mean_token_accuracy": 0.7800502464175224,
|
| 8182 |
+
"num_tokens": 9801263.0,
|
| 8183 |
+
"step": 7370
|
| 8184 |
+
},
|
| 8185 |
+
{
|
| 8186 |
+
"entropy": 0.992620050907135,
|
| 8187 |
+
"epoch": 1.922435688700749,
|
| 8188 |
+
"grad_norm": 0.43336740136146545,
|
| 8189 |
+
"learning_rate": 7.78848658504819e-07,
|
| 8190 |
+
"loss": 0.9871049880981445,
|
| 8191 |
+
"mean_token_accuracy": 0.779982890188694,
|
| 8192 |
+
"num_tokens": 9927368.0,
|
| 8193 |
+
"step": 7380
|
| 8194 |
+
},
|
| 8195 |
+
{
|
| 8196 |
+
"entropy": 1.0000103309750557,
|
| 8197 |
+
"epoch": 1.9250407033539565,
|
| 8198 |
+
"grad_norm": 0.47374773025512695,
|
| 8199 |
+
"learning_rate": 7.528002083876009e-07,
|
| 8200 |
+
"loss": 0.9966364860534668,
|
| 8201 |
+
"mean_token_accuracy": 0.7804308533668518,
|
| 8202 |
+
"num_tokens": 10052811.0,
|
| 8203 |
+
"step": 7390
|
| 8204 |
+
},
|
| 8205 |
+
{
|
| 8206 |
+
"entropy": 0.9543530821800232,
|
| 8207 |
+
"epoch": 1.9276457180071638,
|
| 8208 |
+
"grad_norm": 0.4785780906677246,
|
| 8209 |
+
"learning_rate": 7.26751758270383e-07,
|
| 8210 |
+
"loss": 0.9466802597045898,
|
| 8211 |
+
"mean_token_accuracy": 0.7864332899451256,
|
| 8212 |
+
"num_tokens": 10182309.0,
|
| 8213 |
+
"step": 7400
|
| 8214 |
+
},
|
| 8215 |
+
{
|
| 8216 |
+
"epoch": 1.9276457180071638,
|
| 8217 |
+
"eval_entropy": 0.9832908352869798,
|
| 8218 |
+
"eval_loss": 0.9843752980232239,
|
| 8219 |
+
"eval_mean_token_accuracy": 0.7814531063291077,
|
| 8220 |
+
"eval_num_tokens": 10182309.0,
|
| 8221 |
+
"eval_runtime": 1467.8841,
|
| 8222 |
+
"eval_samples_per_second": 4.637,
|
| 8223 |
+
"eval_steps_per_second": 1.159,
|
| 8224 |
+
"step": 7400
|
| 8225 |
+
},
|
| 8226 |
+
{
|
| 8227 |
+
"entropy": 0.9844714671373367,
|
| 8228 |
+
"epoch": 1.930250732660371,
|
| 8229 |
+
"grad_norm": 0.4658817648887634,
|
| 8230 |
+
"learning_rate": 7.00703308153165e-07,
|
| 8231 |
+
"loss": 0.9865114212036132,
|
| 8232 |
+
"mean_token_accuracy": 0.7815456867218018,
|
| 8233 |
+
"num_tokens": 10306910.0,
|
| 8234 |
+
"step": 7410
|
| 8235 |
+
},
|
| 8236 |
+
{
|
| 8237 |
+
"entropy": 1.0001100838184356,
|
| 8238 |
+
"epoch": 1.9328557473135786,
|
| 8239 |
+
"grad_norm": 0.4151651859283447,
|
| 8240 |
+
"learning_rate": 6.746548580359469e-07,
|
| 8241 |
+
"loss": 0.9890321731567383,
|
| 8242 |
+
"mean_token_accuracy": 0.7787036627531052,
|
| 8243 |
+
"num_tokens": 10433048.0,
|
| 8244 |
+
"step": 7420
|
| 8245 |
+
},
|
| 8246 |
+
{
|
| 8247 |
+
"entropy": 0.9377382323145866,
|
| 8248 |
+
"epoch": 1.935460761966786,
|
| 8249 |
+
"grad_norm": 0.44594910740852356,
|
| 8250 |
+
"learning_rate": 6.486064079187289e-07,
|
| 8251 |
+
"loss": 0.9328557968139648,
|
| 8252 |
+
"mean_token_accuracy": 0.7932988166809082,
|
| 8253 |
+
"num_tokens": 10560426.0,
|
| 8254 |
+
"step": 7430
|
| 8255 |
+
},
|
| 8256 |
+
{
|
| 8257 |
+
"entropy": 0.9599474176764489,
|
| 8258 |
+
"epoch": 1.9380657766199936,
|
| 8259 |
+
"grad_norm": 0.47472286224365234,
|
| 8260 |
+
"learning_rate": 6.225579578015108e-07,
|
| 8261 |
+
"loss": 0.955722427368164,
|
| 8262 |
+
"mean_token_accuracy": 0.7864214271306992,
|
| 8263 |
+
"num_tokens": 10688218.0,
|
| 8264 |
+
"step": 7440
|
| 8265 |
+
},
|
| 8266 |
+
{
|
| 8267 |
+
"entropy": 1.019455748796463,
|
| 8268 |
+
"epoch": 1.9406707912732009,
|
| 8269 |
+
"grad_norm": 0.48136043548583984,
|
| 8270 |
+
"learning_rate": 5.965095076842928e-07,
|
| 8271 |
+
"loss": 1.020223331451416,
|
| 8272 |
+
"mean_token_accuracy": 0.7740492403507233,
|
| 8273 |
+
"num_tokens": 10813588.0,
|
| 8274 |
+
"step": 7450
|
| 8275 |
+
},
|
| 8276 |
+
{
|
| 8277 |
+
"entropy": 0.9933910727500915,
|
| 8278 |
+
"epoch": 1.9432758059264084,
|
| 8279 |
+
"grad_norm": 0.4816676378250122,
|
| 8280 |
+
"learning_rate": 5.704610575670748e-07,
|
| 8281 |
+
"loss": 0.9945768356323242,
|
| 8282 |
+
"mean_token_accuracy": 0.7784866452217102,
|
| 8283 |
+
"num_tokens": 10938414.0,
|
| 8284 |
+
"step": 7460
|
| 8285 |
+
},
|
| 8286 |
+
{
|
| 8287 |
+
"entropy": 0.9734019547700882,
|
| 8288 |
+
"epoch": 1.9458808205796156,
|
| 8289 |
+
"grad_norm": 0.4300379157066345,
|
| 8290 |
+
"learning_rate": 5.444126074498568e-07,
|
| 8291 |
+
"loss": 0.9702705383300781,
|
| 8292 |
+
"mean_token_accuracy": 0.7839167267084122,
|
| 8293 |
+
"num_tokens": 11065293.0,
|
| 8294 |
+
"step": 7470
|
| 8295 |
+
},
|
| 8296 |
+
{
|
| 8297 |
+
"entropy": 1.0072049424052238,
|
| 8298 |
+
"epoch": 1.9484858352328231,
|
| 8299 |
+
"grad_norm": 0.4736027717590332,
|
| 8300 |
+
"learning_rate": 5.183641573326387e-07,
|
| 8301 |
+
"loss": 1.0046442985534667,
|
| 8302 |
+
"mean_token_accuracy": 0.7768969655036926,
|
| 8303 |
+
"num_tokens": 11189583.0,
|
| 8304 |
+
"step": 7480
|
| 8305 |
+
},
|
| 8306 |
+
{
|
| 8307 |
+
"entropy": 0.9816320776939392,
|
| 8308 |
+
"epoch": 1.9510908498860307,
|
| 8309 |
+
"grad_norm": 0.4491042494773865,
|
| 8310 |
+
"learning_rate": 4.923157072154207e-07,
|
| 8311 |
+
"loss": 0.98463134765625,
|
| 8312 |
+
"mean_token_accuracy": 0.7799607574939728,
|
| 8313 |
+
"num_tokens": 11317335.0,
|
| 8314 |
+
"step": 7490
|
| 8315 |
+
},
|
| 8316 |
+
{
|
| 8317 |
+
"entropy": 0.980158931016922,
|
| 8318 |
+
"epoch": 1.9536958645392382,
|
| 8319 |
+
"grad_norm": 0.479579895734787,
|
| 8320 |
+
"learning_rate": 4.662672570982027e-07,
|
| 8321 |
+
"loss": 0.9775474548339844,
|
| 8322 |
+
"mean_token_accuracy": 0.7808455377817154,
|
| 8323 |
+
"num_tokens": 11448281.0,
|
| 8324 |
+
"step": 7500
|
| 8325 |
+
},
|
| 8326 |
+
{
|
| 8327 |
+
"epoch": 1.9536958645392382,
|
| 8328 |
+
"eval_entropy": 0.9845578079007907,
|
| 8329 |
+
"eval_loss": 0.9843409657478333,
|
| 8330 |
+
"eval_mean_token_accuracy": 0.7814459322692364,
|
| 8331 |
+
"eval_num_tokens": 11448281.0,
|
| 8332 |
+
"eval_runtime": 1468.3796,
|
| 8333 |
+
"eval_samples_per_second": 4.635,
|
| 8334 |
+
"eval_steps_per_second": 1.159,
|
| 8335 |
+
"step": 7500
|
| 8336 |
+
},
|
| 8337 |
+
{
|
| 8338 |
+
"entropy": 1.004936684668064,
|
| 8339 |
+
"epoch": 1.9563008791924454,
|
| 8340 |
+
"grad_norm": 0.593953013420105,
|
| 8341 |
+
"learning_rate": 4.4021880698098466e-07,
|
| 8342 |
+
"loss": 1.0079610824584961,
|
| 8343 |
+
"mean_token_accuracy": 0.7764283269643784,
|
| 8344 |
+
"num_tokens": 11576167.0,
|
| 8345 |
+
"step": 7510
|
| 8346 |
+
},
|
| 8347 |
+
{
|
| 8348 |
+
"entropy": 1.018838845193386,
|
| 8349 |
+
"epoch": 1.9589058938456527,
|
| 8350 |
+
"grad_norm": 0.4617692530155182,
|
| 8351 |
+
"learning_rate": 4.1417035686376664e-07,
|
| 8352 |
+
"loss": 1.0125544548034668,
|
| 8353 |
+
"mean_token_accuracy": 0.7748090088367462,
|
| 8354 |
+
"num_tokens": 11704848.0,
|
| 8355 |
+
"step": 7520
|
| 8356 |
+
},
|
| 8357 |
+
{
|
| 8358 |
+
"entropy": 0.9998465582728386,
|
| 8359 |
+
"epoch": 1.9615109084988602,
|
| 8360 |
+
"grad_norm": 0.449637770652771,
|
| 8361 |
+
"learning_rate": 3.8812190674654857e-07,
|
| 8362 |
+
"loss": 1.0013065338134766,
|
| 8363 |
+
"mean_token_accuracy": 0.7779083475470543,
|
| 8364 |
+
"num_tokens": 11832990.0,
|
| 8365 |
+
"step": 7530
|
| 8366 |
+
},
|
| 8367 |
+
{
|
| 8368 |
+
"entropy": 0.9884437397122383,
|
| 8369 |
+
"epoch": 1.9641159231520677,
|
| 8370 |
+
"grad_norm": 0.4621768891811371,
|
| 8371 |
+
"learning_rate": 3.620734566293306e-07,
|
| 8372 |
+
"loss": 0.9889906883239746,
|
| 8373 |
+
"mean_token_accuracy": 0.7803315564990043,
|
| 8374 |
+
"num_tokens": 11959641.0,
|
| 8375 |
+
"step": 7540
|
| 8376 |
+
},
|
| 8377 |
+
{
|
| 8378 |
+
"entropy": 1.0180790185928346,
|
| 8379 |
+
"epoch": 1.9667209378052752,
|
| 8380 |
+
"grad_norm": 0.5363248586654663,
|
| 8381 |
+
"learning_rate": 3.3602500651211253e-07,
|
| 8382 |
+
"loss": 1.0175861358642577,
|
| 8383 |
+
"mean_token_accuracy": 0.7747619211673736,
|
| 8384 |
+
"num_tokens": 12083769.0,
|
| 8385 |
+
"step": 7550
|
| 8386 |
+
},
|
| 8387 |
+
{
|
| 8388 |
+
"entropy": 1.0099502310156823,
|
| 8389 |
+
"epoch": 1.9693259524584827,
|
| 8390 |
+
"grad_norm": 0.4451800286769867,
|
| 8391 |
+
"learning_rate": 3.099765563948945e-07,
|
| 8392 |
+
"loss": 1.0048826217651368,
|
| 8393 |
+
"mean_token_accuracy": 0.7778500929474831,
|
| 8394 |
+
"num_tokens": 12212432.0,
|
| 8395 |
+
"step": 7560
|
| 8396 |
+
},
|
| 8397 |
+
{
|
| 8398 |
+
"entropy": 0.9801075652241706,
|
| 8399 |
+
"epoch": 1.97193096711169,
|
| 8400 |
+
"grad_norm": 0.4810684621334076,
|
| 8401 |
+
"learning_rate": 2.839281062776765e-07,
|
| 8402 |
+
"loss": 0.9807418823242188,
|
| 8403 |
+
"mean_token_accuracy": 0.7826617181301116,
|
| 8404 |
+
"num_tokens": 12338994.0,
|
| 8405 |
+
"step": 7570
|
| 8406 |
+
},
|
| 8407 |
+
{
|
| 8408 |
+
"entropy": 0.979700967669487,
|
| 8409 |
+
"epoch": 1.9745359817648973,
|
| 8410 |
+
"grad_norm": 0.47418564558029175,
|
| 8411 |
+
"learning_rate": 2.578796561604585e-07,
|
| 8412 |
+
"loss": 0.9782838821411133,
|
| 8413 |
+
"mean_token_accuracy": 0.781965833902359,
|
| 8414 |
+
"num_tokens": 12465086.0,
|
| 8415 |
+
"step": 7580
|
| 8416 |
+
},
|
| 8417 |
+
{
|
| 8418 |
+
"entropy": 0.98379997164011,
|
| 8419 |
+
"epoch": 1.9771409964181048,
|
| 8420 |
+
"grad_norm": 0.46422550082206726,
|
| 8421 |
+
"learning_rate": 2.3183120604324046e-07,
|
| 8422 |
+
"loss": 0.9785320281982421,
|
| 8423 |
+
"mean_token_accuracy": 0.7818248748779297,
|
| 8424 |
+
"num_tokens": 12592895.0,
|
| 8425 |
+
"step": 7590
|
| 8426 |
+
},
|
| 8427 |
+
{
|
| 8428 |
+
"entropy": 0.9972235560417175,
|
| 8429 |
+
"epoch": 1.9797460110713123,
|
| 8430 |
+
"grad_norm": 0.4623911380767822,
|
| 8431 |
+
"learning_rate": 2.0578275592602244e-07,
|
| 8432 |
+
"loss": 0.98619384765625,
|
| 8433 |
+
"mean_token_accuracy": 0.780545923113823,
|
| 8434 |
+
"num_tokens": 12714907.0,
|
| 8435 |
+
"step": 7600
|
| 8436 |
+
},
|
| 8437 |
+
{
|
| 8438 |
+
"epoch": 1.9797460110713123,
|
| 8439 |
+
"eval_entropy": 0.9846138954512801,
|
| 8440 |
+
"eval_loss": 0.9843295216560364,
|
| 8441 |
+
"eval_mean_token_accuracy": 0.7814434662058547,
|
| 8442 |
+
"eval_num_tokens": 12714907.0,
|
| 8443 |
+
"eval_runtime": 1468.8159,
|
| 8444 |
+
"eval_samples_per_second": 4.634,
|
| 8445 |
+
"eval_steps_per_second": 1.159,
|
| 8446 |
+
"step": 7600
|
| 8447 |
+
},
|
| 8448 |
+
{
|
| 8449 |
+
"entropy": 0.9750056296586991,
|
| 8450 |
+
"epoch": 1.9823510257245198,
|
| 8451 |
+
"grad_norm": 0.46623268723487854,
|
| 8452 |
+
"learning_rate": 1.7973430580880437e-07,
|
| 8453 |
+
"loss": 0.9689687728881836,
|
| 8454 |
+
"mean_token_accuracy": 0.7827307119965553,
|
| 8455 |
+
"num_tokens": 12841336.0,
|
| 8456 |
+
"step": 7610
|
| 8457 |
+
},
|
| 8458 |
+
{
|
| 8459 |
+
"entropy": 1.0218480482697487,
|
| 8460 |
+
"epoch": 1.984956040377727,
|
| 8461 |
+
"grad_norm": 0.44678768515586853,
|
| 8462 |
+
"learning_rate": 1.5368585569158638e-07,
|
| 8463 |
+
"loss": 1.0249201774597168,
|
| 8464 |
+
"mean_token_accuracy": 0.7761656075716019,
|
| 8465 |
+
"num_tokens": 12970060.0,
|
| 8466 |
+
"step": 7620
|
| 8467 |
+
},
|
| 8468 |
+
{
|
| 8469 |
+
"entropy": 1.0002001956105233,
|
| 8470 |
+
"epoch": 1.9875610550309344,
|
| 8471 |
+
"grad_norm": 0.44759881496429443,
|
| 8472 |
+
"learning_rate": 1.2763740557436834e-07,
|
| 8473 |
+
"loss": 1.000049114227295,
|
| 8474 |
+
"mean_token_accuracy": 0.7782543018460274,
|
| 8475 |
+
"num_tokens": 13096938.0,
|
| 8476 |
+
"step": 7630
|
| 8477 |
+
},
|
| 8478 |
+
{
|
| 8479 |
+
"entropy": 0.9838123112916947,
|
| 8480 |
+
"epoch": 1.990166069684142,
|
| 8481 |
+
"grad_norm": 0.4666384160518646,
|
| 8482 |
+
"learning_rate": 1.015889554571503e-07,
|
| 8483 |
+
"loss": 0.9788585662841797,
|
| 8484 |
+
"mean_token_accuracy": 0.7811343029141427,
|
| 8485 |
+
"num_tokens": 13224908.0,
|
| 8486 |
+
"step": 7640
|
| 8487 |
+
},
|
| 8488 |
+
{
|
| 8489 |
+
"entropy": 1.0147129252552987,
|
| 8490 |
+
"epoch": 1.9927710843373494,
|
| 8491 |
+
"grad_norm": 0.48312798142433167,
|
| 8492 |
+
"learning_rate": 7.554050533993229e-08,
|
| 8493 |
+
"loss": 1.0055107116699218,
|
| 8494 |
+
"mean_token_accuracy": 0.7765903919935226,
|
| 8495 |
+
"num_tokens": 13348971.0,
|
| 8496 |
+
"step": 7650
|
| 8497 |
+
},
|
| 8498 |
+
{
|
| 8499 |
+
"entropy": 1.0030160337686538,
|
| 8500 |
+
"epoch": 1.995376098990557,
|
| 8501 |
+
"grad_norm": 0.4915446639060974,
|
| 8502 |
+
"learning_rate": 4.949205522271425e-08,
|
| 8503 |
+
"loss": 0.9996049880981446,
|
| 8504 |
+
"mean_token_accuracy": 0.7775616005063057,
|
| 8505 |
+
"num_tokens": 13477379.0,
|
| 8506 |
+
"step": 7660
|
| 8507 |
+
},
|
| 8508 |
+
{
|
| 8509 |
+
"entropy": 1.0086737051606178,
|
| 8510 |
+
"epoch": 1.9979811136437644,
|
| 8511 |
+
"grad_norm": 0.42526480555534363,
|
| 8512 |
+
"learning_rate": 2.3443605105496224e-08,
|
| 8513 |
+
"loss": 1.0082104682922364,
|
| 8514 |
+
"mean_token_accuracy": 0.7777216598391533,
|
| 8515 |
+
"num_tokens": 13605551.0,
|
| 8516 |
+
"step": 7670
|
| 8517 |
+
},
|
| 8518 |
+
{
|
| 8519 |
+
"epoch": 2.0,
|
| 8520 |
+
"eval_entropy": 0.9844893059144989,
|
| 8521 |
+
"eval_loss": 0.9843270182609558,
|
| 8522 |
+
"eval_mean_token_accuracy": 0.7814202638196329,
|
| 8523 |
+
"eval_num_tokens": 13701302.0,
|
| 8524 |
+
"eval_runtime": 1467.4042,
|
| 8525 |
+
"eval_samples_per_second": 4.638,
|
| 8526 |
+
"eval_steps_per_second": 1.16,
|
| 8527 |
+
"step": 7678
|
| 8528 |
}
|
| 8529 |
],
|
| 8530 |
"logging_steps": 10,
|
|
|
|
| 8539 |
"should_evaluate": false,
|
| 8540 |
"should_log": false,
|
| 8541 |
"should_save": true,
|
| 8542 |
+
"should_training_stop": true
|
| 8543 |
},
|
| 8544 |
"attributes": {}
|
| 8545 |
}
|
| 8546 |
},
|
| 8547 |
+
"total_flos": 1.2236164104800174e+18,
|
| 8548 |
"train_batch_size": 2,
|
| 8549 |
"trial_name": null,
|
| 8550 |
"trial_params": null
|