Training in progress, step 5806, checkpoint
Browse files- last-checkpoint/adapter_config.json +3 -3
- last-checkpoint/adapter_model.safetensors +1 -1
- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/scaler.pt +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +1125 -4
last-checkpoint/adapter_config.json
CHANGED
|
@@ -31,10 +31,10 @@
|
|
| 31 |
"rank_pattern": {},
|
| 32 |
"revision": null,
|
| 33 |
"target_modules": [
|
| 34 |
-
"
|
| 35 |
-
"v_proj",
|
| 36 |
"o_proj",
|
| 37 |
-
"
|
|
|
|
| 38 |
],
|
| 39 |
"target_parameters": null,
|
| 40 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 31 |
"rank_pattern": {},
|
| 32 |
"revision": null,
|
| 33 |
"target_modules": [
|
| 34 |
+
"k_proj",
|
|
|
|
| 35 |
"o_proj",
|
| 36 |
+
"v_proj",
|
| 37 |
+
"q_proj"
|
| 38 |
],
|
| 39 |
"target_parameters": null,
|
| 40 |
"task_type": "CAUSAL_LM",
|
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 sha256:6f22e14a21fbc01535d8b474a37f7897261e0e9d44ac6c70f5ae6599b1e461d7
|
| 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:c2b171d33aa6f98342456fab464bee21a9699e54d6bfd82c63c9af6ea6adbb40
|
| 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 sha256:38930553ed957b1463a36a308942c5610692c94a54bf3656a9d7406bbfa9c1e1
|
| 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 sha256:497da62cc96f57ba96cd0fc26ae3953441e4ad7c8b6f524685215da682c18652
|
| 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 sha256:0861073ade655d14fb72097d362c2e813ed5d5faa50345ce5bc8f77fc935cfd8
|
| 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:aa80ce769d9e6a0fc0b7fe734e59120bbbef8e392f0c0a33bd722e77bb1ee770
|
| 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,
|
|
@@ -5336,6 +5336,1127 @@
|
|
| 5336 |
"eval_samples_per_second": 5.161,
|
| 5337 |
"eval_steps_per_second": 1.291,
|
| 5338 |
"step": 4800
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5339 |
}
|
| 5340 |
],
|
| 5341 |
"logging_steps": 10,
|
|
@@ -5350,12 +6471,12 @@
|
|
| 5350 |
"should_evaluate": false,
|
| 5351 |
"should_log": false,
|
| 5352 |
"should_save": true,
|
| 5353 |
-
"should_training_stop":
|
| 5354 |
},
|
| 5355 |
"attributes": {}
|
| 5356 |
}
|
| 5357 |
},
|
| 5358 |
-
"total_flos":
|
| 5359 |
"train_batch_size": 2,
|
| 5360 |
"trial_name": null,
|
| 5361 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 5806,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 5336 |
"eval_samples_per_second": 5.161,
|
| 5337 |
"eval_steps_per_second": 1.291,
|
| 5338 |
"step": 4800
|
| 5339 |
+
},
|
| 5340 |
+
{
|
| 5341 |
+
"entropy": 0.857286661863327,
|
| 5342 |
+
"epoch": 0.8284889979761444,
|
| 5343 |
+
"grad_norm": 0.4545460641384125,
|
| 5344 |
+
"learning_rate": 3.434378229417844e-06,
|
| 5345 |
+
"loss": 0.8539424896240234,
|
| 5346 |
+
"mean_token_accuracy": 0.8060515314340592,
|
| 5347 |
+
"num_tokens": 136995.0,
|
| 5348 |
+
"step": 4810
|
| 5349 |
+
},
|
| 5350 |
+
{
|
| 5351 |
+
"entropy": 0.8996703952550889,
|
| 5352 |
+
"epoch": 0.8302114283253671,
|
| 5353 |
+
"grad_norm": 0.43648988008499146,
|
| 5354 |
+
"learning_rate": 3.39993110575267e-06,
|
| 5355 |
+
"loss": 0.8872912406921387,
|
| 5356 |
+
"mean_token_accuracy": 0.8003117755055428,
|
| 5357 |
+
"num_tokens": 269371.0,
|
| 5358 |
+
"step": 4820
|
| 5359 |
+
},
|
| 5360 |
+
{
|
| 5361 |
+
"entropy": 0.897844985127449,
|
| 5362 |
+
"epoch": 0.8319338586745898,
|
| 5363 |
+
"grad_norm": 0.405977338552475,
|
| 5364 |
+
"learning_rate": 3.365483982087496e-06,
|
| 5365 |
+
"loss": 0.8877438545227051,
|
| 5366 |
+
"mean_token_accuracy": 0.8004682302474976,
|
| 5367 |
+
"num_tokens": 407343.0,
|
| 5368 |
+
"step": 4830
|
| 5369 |
+
},
|
| 5370 |
+
{
|
| 5371 |
+
"entropy": 0.8919987097382546,
|
| 5372 |
+
"epoch": 0.8336562890238126,
|
| 5373 |
+
"grad_norm": 0.43606001138687134,
|
| 5374 |
+
"learning_rate": 3.3310368584223218e-06,
|
| 5375 |
+
"loss": 0.8915199279785156,
|
| 5376 |
+
"mean_token_accuracy": 0.8018717259168625,
|
| 5377 |
+
"num_tokens": 543192.0,
|
| 5378 |
+
"step": 4840
|
| 5379 |
+
},
|
| 5380 |
+
{
|
| 5381 |
+
"entropy": 0.8664562180638313,
|
| 5382 |
+
"epoch": 0.8353787193730353,
|
| 5383 |
+
"grad_norm": 0.4047588109970093,
|
| 5384 |
+
"learning_rate": 3.296589734757148e-06,
|
| 5385 |
+
"loss": 0.8543147087097168,
|
| 5386 |
+
"mean_token_accuracy": 0.8059070542454719,
|
| 5387 |
+
"num_tokens": 678711.0,
|
| 5388 |
+
"step": 4850
|
| 5389 |
+
},
|
| 5390 |
+
{
|
| 5391 |
+
"entropy": 0.9019272044301033,
|
| 5392 |
+
"epoch": 0.8371011497222581,
|
| 5393 |
+
"grad_norm": 0.40798190236091614,
|
| 5394 |
+
"learning_rate": 3.262142611091974e-06,
|
| 5395 |
+
"loss": 0.901366138458252,
|
| 5396 |
+
"mean_token_accuracy": 0.7967334732413291,
|
| 5397 |
+
"num_tokens": 815193.0,
|
| 5398 |
+
"step": 4860
|
| 5399 |
+
},
|
| 5400 |
+
{
|
| 5401 |
+
"entropy": 0.8930811554193496,
|
| 5402 |
+
"epoch": 0.8388235800714808,
|
| 5403 |
+
"grad_norm": 0.44522613286972046,
|
| 5404 |
+
"learning_rate": 3.2276954874268003e-06,
|
| 5405 |
+
"loss": 0.8780466079711914,
|
| 5406 |
+
"mean_token_accuracy": 0.7989543616771698,
|
| 5407 |
+
"num_tokens": 951512.0,
|
| 5408 |
+
"step": 4870
|
| 5409 |
+
},
|
| 5410 |
+
{
|
| 5411 |
+
"entropy": 0.897969713807106,
|
| 5412 |
+
"epoch": 0.8405460104207036,
|
| 5413 |
+
"grad_norm": 0.4132952094078064,
|
| 5414 |
+
"learning_rate": 3.193248363761626e-06,
|
| 5415 |
+
"loss": 0.89752197265625,
|
| 5416 |
+
"mean_token_accuracy": 0.7986764028668404,
|
| 5417 |
+
"num_tokens": 1091226.0,
|
| 5418 |
+
"step": 4880
|
| 5419 |
+
},
|
| 5420 |
+
{
|
| 5421 |
+
"entropy": 0.8763776004314423,
|
| 5422 |
+
"epoch": 0.8422684407699264,
|
| 5423 |
+
"grad_norm": 0.464271605014801,
|
| 5424 |
+
"learning_rate": 3.1588012400964525e-06,
|
| 5425 |
+
"loss": 0.866519546508789,
|
| 5426 |
+
"mean_token_accuracy": 0.8051836535334587,
|
| 5427 |
+
"num_tokens": 1230839.0,
|
| 5428 |
+
"step": 4890
|
| 5429 |
+
},
|
| 5430 |
+
{
|
| 5431 |
+
"entropy": 0.8817414686083793,
|
| 5432 |
+
"epoch": 0.8439908711191492,
|
| 5433 |
+
"grad_norm": 0.40158164501190186,
|
| 5434 |
+
"learning_rate": 3.1243541164312784e-06,
|
| 5435 |
+
"loss": 0.8720556259155273,
|
| 5436 |
+
"mean_token_accuracy": 0.8009096115827561,
|
| 5437 |
+
"num_tokens": 1371321.0,
|
| 5438 |
+
"step": 4900
|
| 5439 |
+
},
|
| 5440 |
+
{
|
| 5441 |
+
"epoch": 0.8439908711191492,
|
| 5442 |
+
"eval_entropy": 0.8905009640159186,
|
| 5443 |
+
"eval_loss": 0.886862576007843,
|
| 5444 |
+
"eval_mean_token_accuracy": 0.8004815642430033,
|
| 5445 |
+
"eval_num_tokens": 1371321.0,
|
| 5446 |
+
"eval_runtime": 2214.5743,
|
| 5447 |
+
"eval_samples_per_second": 4.66,
|
| 5448 |
+
"eval_steps_per_second": 1.165,
|
| 5449 |
+
"step": 4900
|
| 5450 |
+
},
|
| 5451 |
+
{
|
| 5452 |
+
"entropy": 0.8792809963226318,
|
| 5453 |
+
"epoch": 0.8457133014683719,
|
| 5454 |
+
"grad_norm": 0.45575663447380066,
|
| 5455 |
+
"learning_rate": 3.0899069927661042e-06,
|
| 5456 |
+
"loss": 0.8713504791259765,
|
| 5457 |
+
"mean_token_accuracy": 0.8019704148173332,
|
| 5458 |
+
"num_tokens": 1508027.0,
|
| 5459 |
+
"step": 4910
|
| 5460 |
+
},
|
| 5461 |
+
{
|
| 5462 |
+
"entropy": 0.852638853341341,
|
| 5463 |
+
"epoch": 0.8474357318175946,
|
| 5464 |
+
"grad_norm": 0.5255561470985413,
|
| 5465 |
+
"learning_rate": 3.05545986910093e-06,
|
| 5466 |
+
"loss": 0.8388608932495117,
|
| 5467 |
+
"mean_token_accuracy": 0.808800981938839,
|
| 5468 |
+
"num_tokens": 1644684.0,
|
| 5469 |
+
"step": 4920
|
| 5470 |
+
},
|
| 5471 |
+
{
|
| 5472 |
+
"entropy": 0.852977229654789,
|
| 5473 |
+
"epoch": 0.8491581621668174,
|
| 5474 |
+
"grad_norm": 0.43369248509407043,
|
| 5475 |
+
"learning_rate": 3.0210127454357564e-06,
|
| 5476 |
+
"loss": 0.8463091850280762,
|
| 5477 |
+
"mean_token_accuracy": 0.8071771740913392,
|
| 5478 |
+
"num_tokens": 1782176.0,
|
| 5479 |
+
"step": 4930
|
| 5480 |
+
},
|
| 5481 |
+
{
|
| 5482 |
+
"entropy": 0.8610281966626644,
|
| 5483 |
+
"epoch": 0.8508805925160401,
|
| 5484 |
+
"grad_norm": 0.44929254055023193,
|
| 5485 |
+
"learning_rate": 2.9865656217705823e-06,
|
| 5486 |
+
"loss": 0.8570366859436035,
|
| 5487 |
+
"mean_token_accuracy": 0.8059208214282989,
|
| 5488 |
+
"num_tokens": 1920144.0,
|
| 5489 |
+
"step": 4940
|
| 5490 |
+
},
|
| 5491 |
+
{
|
| 5492 |
+
"entropy": 0.8712201908230781,
|
| 5493 |
+
"epoch": 0.8526030228652629,
|
| 5494 |
+
"grad_norm": 0.4670601487159729,
|
| 5495 |
+
"learning_rate": 2.9521184981054086e-06,
|
| 5496 |
+
"loss": 0.8666061401367188,
|
| 5497 |
+
"mean_token_accuracy": 0.8054374232888222,
|
| 5498 |
+
"num_tokens": 2059336.0,
|
| 5499 |
+
"step": 4950
|
| 5500 |
+
},
|
| 5501 |
+
{
|
| 5502 |
+
"entropy": 0.8756440788507461,
|
| 5503 |
+
"epoch": 0.8543254532144856,
|
| 5504 |
+
"grad_norm": 0.41233476996421814,
|
| 5505 |
+
"learning_rate": 2.9176713744402345e-06,
|
| 5506 |
+
"loss": 0.862638282775879,
|
| 5507 |
+
"mean_token_accuracy": 0.8034965068101882,
|
| 5508 |
+
"num_tokens": 2201332.0,
|
| 5509 |
+
"step": 4960
|
| 5510 |
+
},
|
| 5511 |
+
{
|
| 5512 |
+
"entropy": 0.9280423223972321,
|
| 5513 |
+
"epoch": 0.8560478835637084,
|
| 5514 |
+
"grad_norm": 0.37263938784599304,
|
| 5515 |
+
"learning_rate": 2.883224250775061e-06,
|
| 5516 |
+
"loss": 0.923769760131836,
|
| 5517 |
+
"mean_token_accuracy": 0.7912818253040313,
|
| 5518 |
+
"num_tokens": 2339368.0,
|
| 5519 |
+
"step": 4970
|
| 5520 |
+
},
|
| 5521 |
+
{
|
| 5522 |
+
"entropy": 0.8748306304216384,
|
| 5523 |
+
"epoch": 0.8577703139129311,
|
| 5524 |
+
"grad_norm": 0.47100237011909485,
|
| 5525 |
+
"learning_rate": 2.8487771271098863e-06,
|
| 5526 |
+
"loss": 0.8737285614013672,
|
| 5527 |
+
"mean_token_accuracy": 0.8047029227018356,
|
| 5528 |
+
"num_tokens": 2476367.0,
|
| 5529 |
+
"step": 4980
|
| 5530 |
+
},
|
| 5531 |
+
{
|
| 5532 |
+
"entropy": 0.9129098683595658,
|
| 5533 |
+
"epoch": 0.8594927442621539,
|
| 5534 |
+
"grad_norm": 0.528985321521759,
|
| 5535 |
+
"learning_rate": 2.8143300034447126e-06,
|
| 5536 |
+
"loss": 0.9065395355224609,
|
| 5537 |
+
"mean_token_accuracy": 0.7945742771029473,
|
| 5538 |
+
"num_tokens": 2613684.0,
|
| 5539 |
+
"step": 4990
|
| 5540 |
+
},
|
| 5541 |
+
{
|
| 5542 |
+
"entropy": 0.8374424375593662,
|
| 5543 |
+
"epoch": 0.8612151746113766,
|
| 5544 |
+
"grad_norm": 0.5343524217605591,
|
| 5545 |
+
"learning_rate": 2.7798828797795385e-06,
|
| 5546 |
+
"loss": 0.8323655128479004,
|
| 5547 |
+
"mean_token_accuracy": 0.8117164567112922,
|
| 5548 |
+
"num_tokens": 2748511.0,
|
| 5549 |
+
"step": 5000
|
| 5550 |
+
},
|
| 5551 |
+
{
|
| 5552 |
+
"epoch": 0.8612151746113766,
|
| 5553 |
+
"eval_entropy": 0.8905347423907262,
|
| 5554 |
+
"eval_loss": 0.886751651763916,
|
| 5555 |
+
"eval_mean_token_accuracy": 0.8005248703313678,
|
| 5556 |
+
"eval_num_tokens": 2748511.0,
|
| 5557 |
+
"eval_runtime": 2213.3653,
|
| 5558 |
+
"eval_samples_per_second": 4.663,
|
| 5559 |
+
"eval_steps_per_second": 1.166,
|
| 5560 |
+
"step": 5000
|
| 5561 |
+
},
|
| 5562 |
+
{
|
| 5563 |
+
"entropy": 0.9117981031537056,
|
| 5564 |
+
"epoch": 0.8629376049605995,
|
| 5565 |
+
"grad_norm": 0.4343831539154053,
|
| 5566 |
+
"learning_rate": 2.745435756114365e-06,
|
| 5567 |
+
"loss": 0.9022266387939453,
|
| 5568 |
+
"mean_token_accuracy": 0.7972292229533195,
|
| 5569 |
+
"num_tokens": 2880947.0,
|
| 5570 |
+
"step": 5010
|
| 5571 |
+
},
|
| 5572 |
+
{
|
| 5573 |
+
"entropy": 0.9056061550974845,
|
| 5574 |
+
"epoch": 0.8646600353098222,
|
| 5575 |
+
"grad_norm": 0.578609824180603,
|
| 5576 |
+
"learning_rate": 2.7109886324491907e-06,
|
| 5577 |
+
"loss": 0.8953428268432617,
|
| 5578 |
+
"mean_token_accuracy": 0.7994224324822425,
|
| 5579 |
+
"num_tokens": 3018280.0,
|
| 5580 |
+
"step": 5020
|
| 5581 |
+
},
|
| 5582 |
+
{
|
| 5583 |
+
"entropy": 0.9164850175380707,
|
| 5584 |
+
"epoch": 0.8663824656590449,
|
| 5585 |
+
"grad_norm": 0.4623558223247528,
|
| 5586 |
+
"learning_rate": 2.676541508784017e-06,
|
| 5587 |
+
"loss": 0.9010392189025879,
|
| 5588 |
+
"mean_token_accuracy": 0.7940018206834794,
|
| 5589 |
+
"num_tokens": 3151878.0,
|
| 5590 |
+
"step": 5030
|
| 5591 |
+
},
|
| 5592 |
+
{
|
| 5593 |
+
"entropy": 0.9213888213038445,
|
| 5594 |
+
"epoch": 0.8681048960082677,
|
| 5595 |
+
"grad_norm": 0.4534718990325928,
|
| 5596 |
+
"learning_rate": 2.642094385118843e-06,
|
| 5597 |
+
"loss": 0.9162769317626953,
|
| 5598 |
+
"mean_token_accuracy": 0.7945156171917915,
|
| 5599 |
+
"num_tokens": 3289251.0,
|
| 5600 |
+
"step": 5040
|
| 5601 |
+
},
|
| 5602 |
+
{
|
| 5603 |
+
"entropy": 0.9475837752223015,
|
| 5604 |
+
"epoch": 0.8698273263574904,
|
| 5605 |
+
"grad_norm": 0.41711097955703735,
|
| 5606 |
+
"learning_rate": 2.607647261453669e-06,
|
| 5607 |
+
"loss": 0.9355989456176758,
|
| 5608 |
+
"mean_token_accuracy": 0.7877664595842362,
|
| 5609 |
+
"num_tokens": 3422221.0,
|
| 5610 |
+
"step": 5050
|
| 5611 |
+
},
|
| 5612 |
+
{
|
| 5613 |
+
"entropy": 0.859785021841526,
|
| 5614 |
+
"epoch": 0.8715497567067132,
|
| 5615 |
+
"grad_norm": 0.41227588057518005,
|
| 5616 |
+
"learning_rate": 2.5732001377884947e-06,
|
| 5617 |
+
"loss": 0.8524208068847656,
|
| 5618 |
+
"mean_token_accuracy": 0.806971238553524,
|
| 5619 |
+
"num_tokens": 3562863.0,
|
| 5620 |
+
"step": 5060
|
| 5621 |
+
},
|
| 5622 |
+
{
|
| 5623 |
+
"entropy": 0.9029248341917991,
|
| 5624 |
+
"epoch": 0.8732721870559359,
|
| 5625 |
+
"grad_norm": 0.597358226776123,
|
| 5626 |
+
"learning_rate": 2.5387530141233206e-06,
|
| 5627 |
+
"loss": 0.8965885162353515,
|
| 5628 |
+
"mean_token_accuracy": 0.7993085518479347,
|
| 5629 |
+
"num_tokens": 3700006.0,
|
| 5630 |
+
"step": 5070
|
| 5631 |
+
},
|
| 5632 |
+
{
|
| 5633 |
+
"entropy": 0.8976858526468277,
|
| 5634 |
+
"epoch": 0.8749946174051587,
|
| 5635 |
+
"grad_norm": 0.49744129180908203,
|
| 5636 |
+
"learning_rate": 2.504305890458147e-06,
|
| 5637 |
+
"loss": 0.8925239562988281,
|
| 5638 |
+
"mean_token_accuracy": 0.7995666921138763,
|
| 5639 |
+
"num_tokens": 3839690.0,
|
| 5640 |
+
"step": 5080
|
| 5641 |
+
},
|
| 5642 |
+
{
|
| 5643 |
+
"entropy": 0.9567562997341156,
|
| 5644 |
+
"epoch": 0.8767170477543814,
|
| 5645 |
+
"grad_norm": 0.4290732443332672,
|
| 5646 |
+
"learning_rate": 2.469858766792973e-06,
|
| 5647 |
+
"loss": 0.9521247863769531,
|
| 5648 |
+
"mean_token_accuracy": 0.7859670892357826,
|
| 5649 |
+
"num_tokens": 3972939.0,
|
| 5650 |
+
"step": 5090
|
| 5651 |
+
},
|
| 5652 |
+
{
|
| 5653 |
+
"entropy": 0.9057339891791344,
|
| 5654 |
+
"epoch": 0.8784394781036042,
|
| 5655 |
+
"grad_norm": 0.4539242684841156,
|
| 5656 |
+
"learning_rate": 2.435411643127799e-06,
|
| 5657 |
+
"loss": 0.89413480758667,
|
| 5658 |
+
"mean_token_accuracy": 0.7966278985142707,
|
| 5659 |
+
"num_tokens": 4106327.0,
|
| 5660 |
+
"step": 5100
|
| 5661 |
+
},
|
| 5662 |
+
{
|
| 5663 |
+
"epoch": 0.8784394781036042,
|
| 5664 |
+
"eval_entropy": 0.889962231752624,
|
| 5665 |
+
"eval_loss": 0.8865664005279541,
|
| 5666 |
+
"eval_mean_token_accuracy": 0.8005476904811438,
|
| 5667 |
+
"eval_num_tokens": 4106327.0,
|
| 5668 |
+
"eval_runtime": 2216.2675,
|
| 5669 |
+
"eval_samples_per_second": 4.657,
|
| 5670 |
+
"eval_steps_per_second": 1.165,
|
| 5671 |
+
"step": 5100
|
| 5672 |
+
},
|
| 5673 |
+
{
|
| 5674 |
+
"entropy": 0.8694351434707641,
|
| 5675 |
+
"epoch": 0.8801619084528269,
|
| 5676 |
+
"grad_norm": 0.4723507761955261,
|
| 5677 |
+
"learning_rate": 2.400964519462625e-06,
|
| 5678 |
+
"loss": 0.8546782493591308,
|
| 5679 |
+
"mean_token_accuracy": 0.8053493618965148,
|
| 5680 |
+
"num_tokens": 4247505.0,
|
| 5681 |
+
"step": 5110
|
| 5682 |
+
},
|
| 5683 |
+
{
|
| 5684 |
+
"entropy": 0.8673742339015007,
|
| 5685 |
+
"epoch": 0.8818843388020496,
|
| 5686 |
+
"grad_norm": 0.4222690463066101,
|
| 5687 |
+
"learning_rate": 2.3665173957974513e-06,
|
| 5688 |
+
"loss": 0.8601169586181641,
|
| 5689 |
+
"mean_token_accuracy": 0.8059160619974136,
|
| 5690 |
+
"num_tokens": 4381996.0,
|
| 5691 |
+
"step": 5120
|
| 5692 |
+
},
|
| 5693 |
+
{
|
| 5694 |
+
"entropy": 0.8701221957802773,
|
| 5695 |
+
"epoch": 0.8836067691512725,
|
| 5696 |
+
"grad_norm": 0.365893691778183,
|
| 5697 |
+
"learning_rate": 2.332070272132277e-06,
|
| 5698 |
+
"loss": 0.8605199813842773,
|
| 5699 |
+
"mean_token_accuracy": 0.8050397902727127,
|
| 5700 |
+
"num_tokens": 4519936.0,
|
| 5701 |
+
"step": 5130
|
| 5702 |
+
},
|
| 5703 |
+
{
|
| 5704 |
+
"entropy": 0.9002679541707039,
|
| 5705 |
+
"epoch": 0.8853291995004952,
|
| 5706 |
+
"grad_norm": 0.41854503750801086,
|
| 5707 |
+
"learning_rate": 2.2976231484671035e-06,
|
| 5708 |
+
"loss": 0.8952171325683593,
|
| 5709 |
+
"mean_token_accuracy": 0.7985835552215577,
|
| 5710 |
+
"num_tokens": 4657396.0,
|
| 5711 |
+
"step": 5140
|
| 5712 |
+
},
|
| 5713 |
+
{
|
| 5714 |
+
"entropy": 0.8730725973844529,
|
| 5715 |
+
"epoch": 0.887051629849718,
|
| 5716 |
+
"grad_norm": 0.4184947907924652,
|
| 5717 |
+
"learning_rate": 2.263176024801929e-06,
|
| 5718 |
+
"loss": 0.874971866607666,
|
| 5719 |
+
"mean_token_accuracy": 0.8019029393792152,
|
| 5720 |
+
"num_tokens": 4793630.0,
|
| 5721 |
+
"step": 5150
|
| 5722 |
+
},
|
| 5723 |
+
{
|
| 5724 |
+
"entropy": 0.8665931671857834,
|
| 5725 |
+
"epoch": 0.8887740601989407,
|
| 5726 |
+
"grad_norm": 0.415822297334671,
|
| 5727 |
+
"learning_rate": 2.2287289011367552e-06,
|
| 5728 |
+
"loss": 0.8619186401367187,
|
| 5729 |
+
"mean_token_accuracy": 0.8056416228413582,
|
| 5730 |
+
"num_tokens": 4932856.0,
|
| 5731 |
+
"step": 5160
|
| 5732 |
+
},
|
| 5733 |
+
{
|
| 5734 |
+
"entropy": 0.8457090623676777,
|
| 5735 |
+
"epoch": 0.8904964905481635,
|
| 5736 |
+
"grad_norm": 0.5788813829421997,
|
| 5737 |
+
"learning_rate": 2.1942817774715815e-06,
|
| 5738 |
+
"loss": 0.8397070884704589,
|
| 5739 |
+
"mean_token_accuracy": 0.8103364869952202,
|
| 5740 |
+
"num_tokens": 5069138.0,
|
| 5741 |
+
"step": 5170
|
| 5742 |
+
},
|
| 5743 |
+
{
|
| 5744 |
+
"entropy": 0.8679139018058777,
|
| 5745 |
+
"epoch": 0.8922189208973862,
|
| 5746 |
+
"grad_norm": 0.4427867531776428,
|
| 5747 |
+
"learning_rate": 2.1598346538064074e-06,
|
| 5748 |
+
"loss": 0.8484273910522461,
|
| 5749 |
+
"mean_token_accuracy": 0.8065137073397637,
|
| 5750 |
+
"num_tokens": 5209109.0,
|
| 5751 |
+
"step": 5180
|
| 5752 |
+
},
|
| 5753 |
+
{
|
| 5754 |
+
"entropy": 0.8567999623715877,
|
| 5755 |
+
"epoch": 0.893941351246609,
|
| 5756 |
+
"grad_norm": 0.4930930435657501,
|
| 5757 |
+
"learning_rate": 2.1253875301412333e-06,
|
| 5758 |
+
"loss": 0.8500759124755859,
|
| 5759 |
+
"mean_token_accuracy": 0.8062406674027442,
|
| 5760 |
+
"num_tokens": 5347756.0,
|
| 5761 |
+
"step": 5190
|
| 5762 |
+
},
|
| 5763 |
+
{
|
| 5764 |
+
"entropy": 0.9153756201267242,
|
| 5765 |
+
"epoch": 0.8956637815958317,
|
| 5766 |
+
"grad_norm": 0.5475414395332336,
|
| 5767 |
+
"learning_rate": 2.090940406476059e-06,
|
| 5768 |
+
"loss": 0.913720703125,
|
| 5769 |
+
"mean_token_accuracy": 0.798624849319458,
|
| 5770 |
+
"num_tokens": 5484156.0,
|
| 5771 |
+
"step": 5200
|
| 5772 |
+
},
|
| 5773 |
+
{
|
| 5774 |
+
"epoch": 0.8956637815958317,
|
| 5775 |
+
"eval_entropy": 0.8903410844865671,
|
| 5776 |
+
"eval_loss": 0.8864830732345581,
|
| 5777 |
+
"eval_mean_token_accuracy": 0.8005986783865671,
|
| 5778 |
+
"eval_num_tokens": 5484156.0,
|
| 5779 |
+
"eval_runtime": 2199.5544,
|
| 5780 |
+
"eval_samples_per_second": 4.692,
|
| 5781 |
+
"eval_steps_per_second": 1.173,
|
| 5782 |
+
"step": 5200
|
| 5783 |
+
},
|
| 5784 |
+
{
|
| 5785 |
+
"entropy": 0.8756707474589348,
|
| 5786 |
+
"epoch": 0.8973862119450545,
|
| 5787 |
+
"grad_norm": 0.4377118945121765,
|
| 5788 |
+
"learning_rate": 2.0564932828108855e-06,
|
| 5789 |
+
"loss": 0.8742427825927734,
|
| 5790 |
+
"mean_token_accuracy": 0.8020416736602783,
|
| 5791 |
+
"num_tokens": 5620110.0,
|
| 5792 |
+
"step": 5210
|
| 5793 |
+
},
|
| 5794 |
+
{
|
| 5795 |
+
"entropy": 0.9009493932127952,
|
| 5796 |
+
"epoch": 0.8991086422942772,
|
| 5797 |
+
"grad_norm": 0.385874480009079,
|
| 5798 |
+
"learning_rate": 2.022046159145712e-06,
|
| 5799 |
+
"loss": 0.9016963958740234,
|
| 5800 |
+
"mean_token_accuracy": 0.7976742312312126,
|
| 5801 |
+
"num_tokens": 5756362.0,
|
| 5802 |
+
"step": 5220
|
| 5803 |
+
},
|
| 5804 |
+
{
|
| 5805 |
+
"entropy": 0.8948132812976837,
|
| 5806 |
+
"epoch": 0.9008310726434999,
|
| 5807 |
+
"grad_norm": 0.4303382337093353,
|
| 5808 |
+
"learning_rate": 1.9875990354805373e-06,
|
| 5809 |
+
"loss": 0.8826349258422852,
|
| 5810 |
+
"mean_token_accuracy": 0.7993564903736115,
|
| 5811 |
+
"num_tokens": 5893634.0,
|
| 5812 |
+
"step": 5230
|
| 5813 |
+
},
|
| 5814 |
+
{
|
| 5815 |
+
"entropy": 0.9013346582651138,
|
| 5816 |
+
"epoch": 0.9025535029927227,
|
| 5817 |
+
"grad_norm": 0.43458035588264465,
|
| 5818 |
+
"learning_rate": 1.9531519118153636e-06,
|
| 5819 |
+
"loss": 0.8878791809082032,
|
| 5820 |
+
"mean_token_accuracy": 0.7967622518539429,
|
| 5821 |
+
"num_tokens": 6031714.0,
|
| 5822 |
+
"step": 5240
|
| 5823 |
+
},
|
| 5824 |
+
{
|
| 5825 |
+
"entropy": 0.900329303741455,
|
| 5826 |
+
"epoch": 0.9042759333419454,
|
| 5827 |
+
"grad_norm": 0.38891515135765076,
|
| 5828 |
+
"learning_rate": 1.9187047881501895e-06,
|
| 5829 |
+
"loss": 0.8943260192871094,
|
| 5830 |
+
"mean_token_accuracy": 0.7962967336177826,
|
| 5831 |
+
"num_tokens": 6167156.0,
|
| 5832 |
+
"step": 5250
|
| 5833 |
+
},
|
| 5834 |
+
{
|
| 5835 |
+
"entropy": 0.8946349762380124,
|
| 5836 |
+
"epoch": 0.9059983636911683,
|
| 5837 |
+
"grad_norm": 0.46829134225845337,
|
| 5838 |
+
"learning_rate": 1.8842576644850158e-06,
|
| 5839 |
+
"loss": 0.8836095809936524,
|
| 5840 |
+
"mean_token_accuracy": 0.7979741424322129,
|
| 5841 |
+
"num_tokens": 6304394.0,
|
| 5842 |
+
"step": 5260
|
| 5843 |
+
},
|
| 5844 |
+
{
|
| 5845 |
+
"entropy": 0.861349032074213,
|
| 5846 |
+
"epoch": 0.907720794040391,
|
| 5847 |
+
"grad_norm": 0.44049137830734253,
|
| 5848 |
+
"learning_rate": 1.8498105408198417e-06,
|
| 5849 |
+
"loss": 0.8538787841796875,
|
| 5850 |
+
"mean_token_accuracy": 0.8079524412751198,
|
| 5851 |
+
"num_tokens": 6445581.0,
|
| 5852 |
+
"step": 5270
|
| 5853 |
+
},
|
| 5854 |
+
{
|
| 5855 |
+
"entropy": 0.8150664329528808,
|
| 5856 |
+
"epoch": 0.9094432243896138,
|
| 5857 |
+
"grad_norm": 0.41160744428634644,
|
| 5858 |
+
"learning_rate": 1.8153634171546678e-06,
|
| 5859 |
+
"loss": 0.8064168930053711,
|
| 5860 |
+
"mean_token_accuracy": 0.8146737694740296,
|
| 5861 |
+
"num_tokens": 6586676.0,
|
| 5862 |
+
"step": 5280
|
| 5863 |
+
},
|
| 5864 |
+
{
|
| 5865 |
+
"entropy": 0.8552966974675655,
|
| 5866 |
+
"epoch": 0.9111656547388365,
|
| 5867 |
+
"grad_norm": 0.4296620190143585,
|
| 5868 |
+
"learning_rate": 1.7809162934894939e-06,
|
| 5869 |
+
"loss": 0.8459642410278321,
|
| 5870 |
+
"mean_token_accuracy": 0.8074897557497025,
|
| 5871 |
+
"num_tokens": 6725692.0,
|
| 5872 |
+
"step": 5290
|
| 5873 |
+
},
|
| 5874 |
+
{
|
| 5875 |
+
"entropy": 0.917242094874382,
|
| 5876 |
+
"epoch": 0.9128880850880593,
|
| 5877 |
+
"grad_norm": 0.4194020926952362,
|
| 5878 |
+
"learning_rate": 1.7464691698243198e-06,
|
| 5879 |
+
"loss": 0.9086408615112305,
|
| 5880 |
+
"mean_token_accuracy": 0.7936558544635772,
|
| 5881 |
+
"num_tokens": 6864077.0,
|
| 5882 |
+
"step": 5300
|
| 5883 |
+
},
|
| 5884 |
+
{
|
| 5885 |
+
"epoch": 0.9128880850880593,
|
| 5886 |
+
"eval_entropy": 0.8895538462104376,
|
| 5887 |
+
"eval_loss": 0.886420488357544,
|
| 5888 |
+
"eval_mean_token_accuracy": 0.8006243253576344,
|
| 5889 |
+
"eval_num_tokens": 6864077.0,
|
| 5890 |
+
"eval_runtime": 2221.5521,
|
| 5891 |
+
"eval_samples_per_second": 4.646,
|
| 5892 |
+
"eval_steps_per_second": 1.162,
|
| 5893 |
+
"step": 5300
|
| 5894 |
+
},
|
| 5895 |
+
{
|
| 5896 |
+
"entropy": 0.8805374369025231,
|
| 5897 |
+
"epoch": 0.914610515437282,
|
| 5898 |
+
"grad_norm": 0.37022849917411804,
|
| 5899 |
+
"learning_rate": 1.7120220461591459e-06,
|
| 5900 |
+
"loss": 0.8718128204345703,
|
| 5901 |
+
"mean_token_accuracy": 0.802679243683815,
|
| 5902 |
+
"num_tokens": 6998242.0,
|
| 5903 |
+
"step": 5310
|
| 5904 |
+
},
|
| 5905 |
+
{
|
| 5906 |
+
"entropy": 0.9066009625792504,
|
| 5907 |
+
"epoch": 0.9163329457865047,
|
| 5908 |
+
"grad_norm": 0.6012776494026184,
|
| 5909 |
+
"learning_rate": 1.677574922493972e-06,
|
| 5910 |
+
"loss": 0.9060131072998047,
|
| 5911 |
+
"mean_token_accuracy": 0.7947992786765099,
|
| 5912 |
+
"num_tokens": 7136912.0,
|
| 5913 |
+
"step": 5320
|
| 5914 |
+
},
|
| 5915 |
+
{
|
| 5916 |
+
"entropy": 0.8422975018620491,
|
| 5917 |
+
"epoch": 0.9180553761357275,
|
| 5918 |
+
"grad_norm": 0.4443066716194153,
|
| 5919 |
+
"learning_rate": 1.643127798828798e-06,
|
| 5920 |
+
"loss": 0.8419797897338868,
|
| 5921 |
+
"mean_token_accuracy": 0.8104378402233123,
|
| 5922 |
+
"num_tokens": 7279298.0,
|
| 5923 |
+
"step": 5330
|
| 5924 |
+
},
|
| 5925 |
+
{
|
| 5926 |
+
"entropy": 0.8393626287579536,
|
| 5927 |
+
"epoch": 0.9197778064849502,
|
| 5928 |
+
"grad_norm": 0.42755192518234253,
|
| 5929 |
+
"learning_rate": 1.608680675163624e-06,
|
| 5930 |
+
"loss": 0.8286300659179687,
|
| 5931 |
+
"mean_token_accuracy": 0.8120000720024109,
|
| 5932 |
+
"num_tokens": 7416488.0,
|
| 5933 |
+
"step": 5340
|
| 5934 |
+
},
|
| 5935 |
+
{
|
| 5936 |
+
"entropy": 0.8612012237310409,
|
| 5937 |
+
"epoch": 0.921500236834173,
|
| 5938 |
+
"grad_norm": 0.42363837361335754,
|
| 5939 |
+
"learning_rate": 1.57423355149845e-06,
|
| 5940 |
+
"loss": 0.8535400390625,
|
| 5941 |
+
"mean_token_accuracy": 0.8065411135554313,
|
| 5942 |
+
"num_tokens": 7555188.0,
|
| 5943 |
+
"step": 5350
|
| 5944 |
+
},
|
| 5945 |
+
{
|
| 5946 |
+
"entropy": 0.8624228790402413,
|
| 5947 |
+
"epoch": 0.9232226671833957,
|
| 5948 |
+
"grad_norm": 0.4368661642074585,
|
| 5949 |
+
"learning_rate": 1.5397864278332761e-06,
|
| 5950 |
+
"loss": 0.8525060653686524,
|
| 5951 |
+
"mean_token_accuracy": 0.8070851415395737,
|
| 5952 |
+
"num_tokens": 7694113.0,
|
| 5953 |
+
"step": 5360
|
| 5954 |
+
},
|
| 5955 |
+
{
|
| 5956 |
+
"entropy": 0.89511788636446,
|
| 5957 |
+
"epoch": 0.9249450975326186,
|
| 5958 |
+
"grad_norm": 0.44434964656829834,
|
| 5959 |
+
"learning_rate": 1.5053393041681022e-06,
|
| 5960 |
+
"loss": 0.8826745033264161,
|
| 5961 |
+
"mean_token_accuracy": 0.8017610549926758,
|
| 5962 |
+
"num_tokens": 7831182.0,
|
| 5963 |
+
"step": 5370
|
| 5964 |
+
},
|
| 5965 |
+
{
|
| 5966 |
+
"entropy": 0.8900723055005073,
|
| 5967 |
+
"epoch": 0.9266675278818413,
|
| 5968 |
+
"grad_norm": 0.3896443545818329,
|
| 5969 |
+
"learning_rate": 1.4708921805029281e-06,
|
| 5970 |
+
"loss": 0.8831830978393554,
|
| 5971 |
+
"mean_token_accuracy": 0.800842498242855,
|
| 5972 |
+
"num_tokens": 7967772.0,
|
| 5973 |
+
"step": 5380
|
| 5974 |
+
},
|
| 5975 |
+
{
|
| 5976 |
+
"entropy": 0.8480780109763145,
|
| 5977 |
+
"epoch": 0.9283899582310641,
|
| 5978 |
+
"grad_norm": 0.47855135798454285,
|
| 5979 |
+
"learning_rate": 1.4364450568377542e-06,
|
| 5980 |
+
"loss": 0.8430942535400391,
|
| 5981 |
+
"mean_token_accuracy": 0.8073634237051011,
|
| 5982 |
+
"num_tokens": 8109699.0,
|
| 5983 |
+
"step": 5390
|
| 5984 |
+
},
|
| 5985 |
+
{
|
| 5986 |
+
"entropy": 0.8448813557624817,
|
| 5987 |
+
"epoch": 0.9301123885802868,
|
| 5988 |
+
"grad_norm": 0.40834853053092957,
|
| 5989 |
+
"learning_rate": 1.4019979331725803e-06,
|
| 5990 |
+
"loss": 0.8393930435180664,
|
| 5991 |
+
"mean_token_accuracy": 0.8103328227996827,
|
| 5992 |
+
"num_tokens": 8247058.0,
|
| 5993 |
+
"step": 5400
|
| 5994 |
+
},
|
| 5995 |
+
{
|
| 5996 |
+
"epoch": 0.9301123885802868,
|
| 5997 |
+
"eval_entropy": 0.8897033281933194,
|
| 5998 |
+
"eval_loss": 0.8863592147827148,
|
| 5999 |
+
"eval_mean_token_accuracy": 0.8006240403592148,
|
| 6000 |
+
"eval_num_tokens": 8247058.0,
|
| 6001 |
+
"eval_runtime": 2208.4615,
|
| 6002 |
+
"eval_samples_per_second": 4.673,
|
| 6003 |
+
"eval_steps_per_second": 1.169,
|
| 6004 |
+
"step": 5400
|
| 6005 |
+
},
|
| 6006 |
+
{
|
| 6007 |
+
"entropy": 0.8857871651649475,
|
| 6008 |
+
"epoch": 0.9318348189295096,
|
| 6009 |
+
"grad_norm": 0.4397316575050354,
|
| 6010 |
+
"learning_rate": 1.3675508095074064e-06,
|
| 6011 |
+
"loss": 0.8728399276733398,
|
| 6012 |
+
"mean_token_accuracy": 0.8018044486641884,
|
| 6013 |
+
"num_tokens": 8384933.0,
|
| 6014 |
+
"step": 5410
|
| 6015 |
+
},
|
| 6016 |
+
{
|
| 6017 |
+
"entropy": 0.9105540379881859,
|
| 6018 |
+
"epoch": 0.9335572492787323,
|
| 6019 |
+
"grad_norm": 0.4545707702636719,
|
| 6020 |
+
"learning_rate": 1.3331036858422323e-06,
|
| 6021 |
+
"loss": 0.9085962295532226,
|
| 6022 |
+
"mean_token_accuracy": 0.795919018983841,
|
| 6023 |
+
"num_tokens": 8526364.0,
|
| 6024 |
+
"step": 5420
|
| 6025 |
+
},
|
| 6026 |
+
{
|
| 6027 |
+
"entropy": 0.8637060999870301,
|
| 6028 |
+
"epoch": 0.935279679627955,
|
| 6029 |
+
"grad_norm": 0.36590105295181274,
|
| 6030 |
+
"learning_rate": 1.2986565621770584e-06,
|
| 6031 |
+
"loss": 0.8607066154479981,
|
| 6032 |
+
"mean_token_accuracy": 0.8060595214366912,
|
| 6033 |
+
"num_tokens": 8664594.0,
|
| 6034 |
+
"step": 5430
|
| 6035 |
+
},
|
| 6036 |
+
{
|
| 6037 |
+
"entropy": 0.9462354958057404,
|
| 6038 |
+
"epoch": 0.9370021099771778,
|
| 6039 |
+
"grad_norm": 0.4243916869163513,
|
| 6040 |
+
"learning_rate": 1.2642094385118845e-06,
|
| 6041 |
+
"loss": 0.9325800895690918,
|
| 6042 |
+
"mean_token_accuracy": 0.7894501313567162,
|
| 6043 |
+
"num_tokens": 8798030.0,
|
| 6044 |
+
"step": 5440
|
| 6045 |
+
},
|
| 6046 |
+
{
|
| 6047 |
+
"entropy": 0.8978266701102257,
|
| 6048 |
+
"epoch": 0.9387245403264005,
|
| 6049 |
+
"grad_norm": 0.4084837734699249,
|
| 6050 |
+
"learning_rate": 1.2297623148467104e-06,
|
| 6051 |
+
"loss": 0.8851606369018554,
|
| 6052 |
+
"mean_token_accuracy": 0.8009923383593559,
|
| 6053 |
+
"num_tokens": 8937218.0,
|
| 6054 |
+
"step": 5450
|
| 6055 |
+
},
|
| 6056 |
+
{
|
| 6057 |
+
"entropy": 0.8797792851924896,
|
| 6058 |
+
"epoch": 0.9404469706756233,
|
| 6059 |
+
"grad_norm": 0.4102013409137726,
|
| 6060 |
+
"learning_rate": 1.1953151911815365e-06,
|
| 6061 |
+
"loss": 0.8767299652099609,
|
| 6062 |
+
"mean_token_accuracy": 0.8016744539141655,
|
| 6063 |
+
"num_tokens": 9075609.0,
|
| 6064 |
+
"step": 5460
|
| 6065 |
+
},
|
| 6066 |
+
{
|
| 6067 |
+
"entropy": 0.877577556669712,
|
| 6068 |
+
"epoch": 0.942169401024846,
|
| 6069 |
+
"grad_norm": 0.40450313687324524,
|
| 6070 |
+
"learning_rate": 1.1608680675163626e-06,
|
| 6071 |
+
"loss": 0.8800636291503906,
|
| 6072 |
+
"mean_token_accuracy": 0.8021259069442749,
|
| 6073 |
+
"num_tokens": 9215492.0,
|
| 6074 |
+
"step": 5470
|
| 6075 |
+
},
|
| 6076 |
+
{
|
| 6077 |
+
"entropy": 0.8801092639565468,
|
| 6078 |
+
"epoch": 0.9438918313740688,
|
| 6079 |
+
"grad_norm": 0.4468408226966858,
|
| 6080 |
+
"learning_rate": 1.1264209438511885e-06,
|
| 6081 |
+
"loss": 0.8776715278625489,
|
| 6082 |
+
"mean_token_accuracy": 0.8040337398648262,
|
| 6083 |
+
"num_tokens": 9355581.0,
|
| 6084 |
+
"step": 5480
|
| 6085 |
+
},
|
| 6086 |
+
{
|
| 6087 |
+
"entropy": 0.8833095215260982,
|
| 6088 |
+
"epoch": 0.9456142617232915,
|
| 6089 |
+
"grad_norm": 0.5369368195533752,
|
| 6090 |
+
"learning_rate": 1.0919738201860146e-06,
|
| 6091 |
+
"loss": 0.8716952323913574,
|
| 6092 |
+
"mean_token_accuracy": 0.8020655795931816,
|
| 6093 |
+
"num_tokens": 9492214.0,
|
| 6094 |
+
"step": 5490
|
| 6095 |
+
},
|
| 6096 |
+
{
|
| 6097 |
+
"entropy": 0.9180077627301216,
|
| 6098 |
+
"epoch": 0.9473366920725144,
|
| 6099 |
+
"grad_norm": 0.4173908829689026,
|
| 6100 |
+
"learning_rate": 1.0575266965208405e-06,
|
| 6101 |
+
"loss": 0.9027196884155273,
|
| 6102 |
+
"mean_token_accuracy": 0.7945086941123009,
|
| 6103 |
+
"num_tokens": 9630657.0,
|
| 6104 |
+
"step": 5500
|
| 6105 |
+
},
|
| 6106 |
+
{
|
| 6107 |
+
"epoch": 0.9473366920725144,
|
| 6108 |
+
"eval_entropy": 0.89089541919515,
|
| 6109 |
+
"eval_loss": 0.8862592577934265,
|
| 6110 |
+
"eval_mean_token_accuracy": 0.8006462655479434,
|
| 6111 |
+
"eval_num_tokens": 9630657.0,
|
| 6112 |
+
"eval_runtime": 2217.3988,
|
| 6113 |
+
"eval_samples_per_second": 4.655,
|
| 6114 |
+
"eval_steps_per_second": 1.164,
|
| 6115 |
+
"step": 5500
|
| 6116 |
+
},
|
| 6117 |
+
{
|
| 6118 |
+
"entropy": 0.9045361965894699,
|
| 6119 |
+
"epoch": 0.9490591224217371,
|
| 6120 |
+
"grad_norm": 0.44494736194610596,
|
| 6121 |
+
"learning_rate": 1.0230795728556668e-06,
|
| 6122 |
+
"loss": 0.9016293525695801,
|
| 6123 |
+
"mean_token_accuracy": 0.7991117894649505,
|
| 6124 |
+
"num_tokens": 9765697.0,
|
| 6125 |
+
"step": 5510
|
| 6126 |
+
},
|
| 6127 |
+
{
|
| 6128 |
+
"entropy": 0.908899100124836,
|
| 6129 |
+
"epoch": 0.9507815527709598,
|
| 6130 |
+
"grad_norm": 0.4863620102405548,
|
| 6131 |
+
"learning_rate": 9.886324491904926e-07,
|
| 6132 |
+
"loss": 0.9006387710571289,
|
| 6133 |
+
"mean_token_accuracy": 0.7961707264184952,
|
| 6134 |
+
"num_tokens": 9902667.0,
|
| 6135 |
+
"step": 5520
|
| 6136 |
+
},
|
| 6137 |
+
{
|
| 6138 |
+
"entropy": 0.9102652281522751,
|
| 6139 |
+
"epoch": 0.9525039831201826,
|
| 6140 |
+
"grad_norm": 0.5516873598098755,
|
| 6141 |
+
"learning_rate": 9.541853255253187e-07,
|
| 6142 |
+
"loss": 0.8945722579956055,
|
| 6143 |
+
"mean_token_accuracy": 0.7974153965711593,
|
| 6144 |
+
"num_tokens": 10038391.0,
|
| 6145 |
+
"step": 5530
|
| 6146 |
+
},
|
| 6147 |
+
{
|
| 6148 |
+
"entropy": 0.8632800906896592,
|
| 6149 |
+
"epoch": 0.9542264134694053,
|
| 6150 |
+
"grad_norm": 0.420607328414917,
|
| 6151 |
+
"learning_rate": 9.197382018601447e-07,
|
| 6152 |
+
"loss": 0.8471515655517579,
|
| 6153 |
+
"mean_token_accuracy": 0.8071994781494141,
|
| 6154 |
+
"num_tokens": 10173686.0,
|
| 6155 |
+
"step": 5540
|
| 6156 |
+
},
|
| 6157 |
+
{
|
| 6158 |
+
"entropy": 0.885202006995678,
|
| 6159 |
+
"epoch": 0.9559488438186281,
|
| 6160 |
+
"grad_norm": 0.4669937789440155,
|
| 6161 |
+
"learning_rate": 8.852910781949708e-07,
|
| 6162 |
+
"loss": 0.886192512512207,
|
| 6163 |
+
"mean_token_accuracy": 0.7994555085897446,
|
| 6164 |
+
"num_tokens": 10311741.0,
|
| 6165 |
+
"step": 5550
|
| 6166 |
+
},
|
| 6167 |
+
{
|
| 6168 |
+
"entropy": 0.8559036374092102,
|
| 6169 |
+
"epoch": 0.9576712741678508,
|
| 6170 |
+
"grad_norm": 0.46770304441452026,
|
| 6171 |
+
"learning_rate": 8.508439545297968e-07,
|
| 6172 |
+
"loss": 0.8477782249450684,
|
| 6173 |
+
"mean_token_accuracy": 0.8067097589373589,
|
| 6174 |
+
"num_tokens": 10453099.0,
|
| 6175 |
+
"step": 5560
|
| 6176 |
+
},
|
| 6177 |
+
{
|
| 6178 |
+
"entropy": 0.9203044310212135,
|
| 6179 |
+
"epoch": 0.9593937045170736,
|
| 6180 |
+
"grad_norm": 0.4336691200733185,
|
| 6181 |
+
"learning_rate": 8.163968308646229e-07,
|
| 6182 |
+
"loss": 0.9105629920959473,
|
| 6183 |
+
"mean_token_accuracy": 0.7944615721702576,
|
| 6184 |
+
"num_tokens": 10591433.0,
|
| 6185 |
+
"step": 5570
|
| 6186 |
+
},
|
| 6187 |
+
{
|
| 6188 |
+
"entropy": 0.8992323987185955,
|
| 6189 |
+
"epoch": 0.9611161348662963,
|
| 6190 |
+
"grad_norm": 0.4164210557937622,
|
| 6191 |
+
"learning_rate": 7.819497071994489e-07,
|
| 6192 |
+
"loss": 0.8976916313171387,
|
| 6193 |
+
"mean_token_accuracy": 0.7977223128080368,
|
| 6194 |
+
"num_tokens": 10728991.0,
|
| 6195 |
+
"step": 5580
|
| 6196 |
+
},
|
| 6197 |
+
{
|
| 6198 |
+
"entropy": 0.8688705429434777,
|
| 6199 |
+
"epoch": 0.9628385652155191,
|
| 6200 |
+
"grad_norm": 0.4670332372188568,
|
| 6201 |
+
"learning_rate": 7.47502583534275e-07,
|
| 6202 |
+
"loss": 0.8597276687622071,
|
| 6203 |
+
"mean_token_accuracy": 0.8061973169445992,
|
| 6204 |
+
"num_tokens": 10867884.0,
|
| 6205 |
+
"step": 5590
|
| 6206 |
+
},
|
| 6207 |
+
{
|
| 6208 |
+
"entropy": 0.9053284600377083,
|
| 6209 |
+
"epoch": 0.9645609955647418,
|
| 6210 |
+
"grad_norm": 0.3938577175140381,
|
| 6211 |
+
"learning_rate": 7.13055459869101e-07,
|
| 6212 |
+
"loss": 0.8982341766357422,
|
| 6213 |
+
"mean_token_accuracy": 0.7990050241351128,
|
| 6214 |
+
"num_tokens": 11001110.0,
|
| 6215 |
+
"step": 5600
|
| 6216 |
+
},
|
| 6217 |
+
{
|
| 6218 |
+
"epoch": 0.9645609955647418,
|
| 6219 |
+
"eval_entropy": 0.8901571277521785,
|
| 6220 |
+
"eval_loss": 0.8862133026123047,
|
| 6221 |
+
"eval_mean_token_accuracy": 0.8006129994590077,
|
| 6222 |
+
"eval_num_tokens": 11001110.0,
|
| 6223 |
+
"eval_runtime": 2221.0741,
|
| 6224 |
+
"eval_samples_per_second": 4.647,
|
| 6225 |
+
"eval_steps_per_second": 1.162,
|
| 6226 |
+
"step": 5600
|
| 6227 |
+
},
|
| 6228 |
+
{
|
| 6229 |
+
"entropy": 0.909729877114296,
|
| 6230 |
+
"epoch": 0.9662834259139647,
|
| 6231 |
+
"grad_norm": NaN,
|
| 6232 |
+
"learning_rate": 6.78608336203927e-07,
|
| 6233 |
+
"loss": 0.9095258712768555,
|
| 6234 |
+
"mean_token_accuracy": 0.8012943014502525,
|
| 6235 |
+
"num_tokens": 11139523.0,
|
| 6236 |
+
"step": 5610
|
| 6237 |
+
},
|
| 6238 |
+
{
|
| 6239 |
+
"entropy": 0.8407525554299354,
|
| 6240 |
+
"epoch": 0.9680058562631874,
|
| 6241 |
+
"grad_norm": 0.43725696206092834,
|
| 6242 |
+
"learning_rate": 6.441612125387531e-07,
|
| 6243 |
+
"loss": 0.8371444702148437,
|
| 6244 |
+
"mean_token_accuracy": 0.810321681201458,
|
| 6245 |
+
"num_tokens": 11276005.0,
|
| 6246 |
+
"step": 5620
|
| 6247 |
+
},
|
| 6248 |
+
{
|
| 6249 |
+
"entropy": 0.901355029642582,
|
| 6250 |
+
"epoch": 0.9697282866124101,
|
| 6251 |
+
"grad_norm": 0.5541648268699646,
|
| 6252 |
+
"learning_rate": 6.097140888735791e-07,
|
| 6253 |
+
"loss": 0.897189712524414,
|
| 6254 |
+
"mean_token_accuracy": 0.7999485448002815,
|
| 6255 |
+
"num_tokens": 11409561.0,
|
| 6256 |
+
"step": 5630
|
| 6257 |
+
},
|
| 6258 |
+
{
|
| 6259 |
+
"entropy": 0.9109830990433693,
|
| 6260 |
+
"epoch": 0.9714507169616329,
|
| 6261 |
+
"grad_norm": 0.5093613266944885,
|
| 6262 |
+
"learning_rate": 5.752669652084052e-07,
|
| 6263 |
+
"loss": 0.9014092445373535,
|
| 6264 |
+
"mean_token_accuracy": 0.7963637366890908,
|
| 6265 |
+
"num_tokens": 11545625.0,
|
| 6266 |
+
"step": 5640
|
| 6267 |
+
},
|
| 6268 |
+
{
|
| 6269 |
+
"entropy": 0.8920518733561039,
|
| 6270 |
+
"epoch": 0.9731731473108556,
|
| 6271 |
+
"grad_norm": 0.43631625175476074,
|
| 6272 |
+
"learning_rate": 5.408198415432312e-07,
|
| 6273 |
+
"loss": 0.8796001434326172,
|
| 6274 |
+
"mean_token_accuracy": 0.7989365607500076,
|
| 6275 |
+
"num_tokens": 11680063.0,
|
| 6276 |
+
"step": 5650
|
| 6277 |
+
},
|
| 6278 |
+
{
|
| 6279 |
+
"entropy": 0.9035694569349288,
|
| 6280 |
+
"epoch": 0.9748955776600784,
|
| 6281 |
+
"grad_norm": 0.40075695514678955,
|
| 6282 |
+
"learning_rate": 5.063727178780573e-07,
|
| 6283 |
+
"loss": 0.8951308250427246,
|
| 6284 |
+
"mean_token_accuracy": 0.7964143276214599,
|
| 6285 |
+
"num_tokens": 11819814.0,
|
| 6286 |
+
"step": 5660
|
| 6287 |
+
},
|
| 6288 |
+
{
|
| 6289 |
+
"entropy": 0.9101156964898109,
|
| 6290 |
+
"epoch": 0.9766180080093011,
|
| 6291 |
+
"grad_norm": 0.3921527564525604,
|
| 6292 |
+
"learning_rate": 4.719255942128833e-07,
|
| 6293 |
+
"loss": 0.9054186820983887,
|
| 6294 |
+
"mean_token_accuracy": 0.7962990015745163,
|
| 6295 |
+
"num_tokens": 11958312.0,
|
| 6296 |
+
"step": 5670
|
| 6297 |
+
},
|
| 6298 |
+
{
|
| 6299 |
+
"entropy": 0.8994546860456467,
|
| 6300 |
+
"epoch": 0.9783404383585239,
|
| 6301 |
+
"grad_norm": 0.4922078549861908,
|
| 6302 |
+
"learning_rate": 4.3747847054770927e-07,
|
| 6303 |
+
"loss": 0.8862943649291992,
|
| 6304 |
+
"mean_token_accuracy": 0.7998211741447449,
|
| 6305 |
+
"num_tokens": 12098750.0,
|
| 6306 |
+
"step": 5680
|
| 6307 |
+
},
|
| 6308 |
+
{
|
| 6309 |
+
"entropy": 0.8943607084453106,
|
| 6310 |
+
"epoch": 0.9800628687077466,
|
| 6311 |
+
"grad_norm": 0.35828521847724915,
|
| 6312 |
+
"learning_rate": 4.030313468825353e-07,
|
| 6313 |
+
"loss": 0.8841873168945312,
|
| 6314 |
+
"mean_token_accuracy": 0.8006264850497246,
|
| 6315 |
+
"num_tokens": 12238206.0,
|
| 6316 |
+
"step": 5690
|
| 6317 |
+
},
|
| 6318 |
+
{
|
| 6319 |
+
"entropy": 0.8610037744045258,
|
| 6320 |
+
"epoch": 0.9817852990569694,
|
| 6321 |
+
"grad_norm": 0.4048604965209961,
|
| 6322 |
+
"learning_rate": 3.6858422321736136e-07,
|
| 6323 |
+
"loss": 0.850681209564209,
|
| 6324 |
+
"mean_token_accuracy": 0.8062293291091919,
|
| 6325 |
+
"num_tokens": 12373855.0,
|
| 6326 |
+
"step": 5700
|
| 6327 |
+
},
|
| 6328 |
+
{
|
| 6329 |
+
"epoch": 0.9817852990569694,
|
| 6330 |
+
"eval_entropy": 0.8902986454021464,
|
| 6331 |
+
"eval_loss": 0.8861925601959229,
|
| 6332 |
+
"eval_mean_token_accuracy": 0.8006364003210648,
|
| 6333 |
+
"eval_num_tokens": 12373855.0,
|
| 6334 |
+
"eval_runtime": 2198.4056,
|
| 6335 |
+
"eval_samples_per_second": 4.695,
|
| 6336 |
+
"eval_steps_per_second": 1.174,
|
| 6337 |
+
"step": 5700
|
| 6338 |
+
},
|
| 6339 |
+
{
|
| 6340 |
+
"entropy": 0.8190450638532638,
|
| 6341 |
+
"epoch": 0.9835077294061921,
|
| 6342 |
+
"grad_norm": 0.46052828431129456,
|
| 6343 |
+
"learning_rate": 3.341370995521874e-07,
|
| 6344 |
+
"loss": 0.8094842910766602,
|
| 6345 |
+
"mean_token_accuracy": 0.8147952750325202,
|
| 6346 |
+
"num_tokens": 12510961.0,
|
| 6347 |
+
"step": 5710
|
| 6348 |
+
},
|
| 6349 |
+
{
|
| 6350 |
+
"entropy": 0.9092415377497673,
|
| 6351 |
+
"epoch": 0.9852301597554148,
|
| 6352 |
+
"grad_norm": 0.46878325939178467,
|
| 6353 |
+
"learning_rate": 2.9968997588701345e-07,
|
| 6354 |
+
"loss": 0.8955598831176758,
|
| 6355 |
+
"mean_token_accuracy": 0.7964127615094185,
|
| 6356 |
+
"num_tokens": 12650748.0,
|
| 6357 |
+
"step": 5720
|
| 6358 |
+
},
|
| 6359 |
+
{
|
| 6360 |
+
"entropy": 0.8547648385167121,
|
| 6361 |
+
"epoch": 0.9869525901046377,
|
| 6362 |
+
"grad_norm": 0.47159335017204285,
|
| 6363 |
+
"learning_rate": 2.652428522218395e-07,
|
| 6364 |
+
"loss": 0.8555225372314453,
|
| 6365 |
+
"mean_token_accuracy": 0.807997515797615,
|
| 6366 |
+
"num_tokens": 12793599.0,
|
| 6367 |
+
"step": 5730
|
| 6368 |
+
},
|
| 6369 |
+
{
|
| 6370 |
+
"entropy": 0.8646257355809212,
|
| 6371 |
+
"epoch": 0.9886750204538604,
|
| 6372 |
+
"grad_norm": 0.48610877990722656,
|
| 6373 |
+
"learning_rate": 2.3079572855666552e-07,
|
| 6374 |
+
"loss": 0.8589855194091797,
|
| 6375 |
+
"mean_token_accuracy": 0.806700712442398,
|
| 6376 |
+
"num_tokens": 12932824.0,
|
| 6377 |
+
"step": 5740
|
| 6378 |
+
},
|
| 6379 |
+
{
|
| 6380 |
+
"entropy": 0.8452193960547447,
|
| 6381 |
+
"epoch": 0.9903974508030832,
|
| 6382 |
+
"grad_norm": 0.4377225637435913,
|
| 6383 |
+
"learning_rate": 1.9634860489149156e-07,
|
| 6384 |
+
"loss": 0.8359296798706055,
|
| 6385 |
+
"mean_token_accuracy": 0.8110934183001518,
|
| 6386 |
+
"num_tokens": 13071288.0,
|
| 6387 |
+
"step": 5750
|
| 6388 |
+
},
|
| 6389 |
+
{
|
| 6390 |
+
"entropy": 0.8705169402062893,
|
| 6391 |
+
"epoch": 0.9921198811523059,
|
| 6392 |
+
"grad_norm": 0.3535923659801483,
|
| 6393 |
+
"learning_rate": 1.619014812263176e-07,
|
| 6394 |
+
"loss": 0.8605812072753907,
|
| 6395 |
+
"mean_token_accuracy": 0.804425074160099,
|
| 6396 |
+
"num_tokens": 13209166.0,
|
| 6397 |
+
"step": 5760
|
| 6398 |
+
},
|
| 6399 |
+
{
|
| 6400 |
+
"entropy": 0.9036453366279602,
|
| 6401 |
+
"epoch": 0.9938423115015287,
|
| 6402 |
+
"grad_norm": 0.38716036081314087,
|
| 6403 |
+
"learning_rate": 1.2745435756114365e-07,
|
| 6404 |
+
"loss": 0.9031167030334473,
|
| 6405 |
+
"mean_token_accuracy": 0.7980457425117493,
|
| 6406 |
+
"num_tokens": 13344912.0,
|
| 6407 |
+
"step": 5770
|
| 6408 |
+
},
|
| 6409 |
+
{
|
| 6410 |
+
"entropy": 0.8937770411372185,
|
| 6411 |
+
"epoch": 0.9955647418507514,
|
| 6412 |
+
"grad_norm": 0.442220002412796,
|
| 6413 |
+
"learning_rate": 9.300723389596969e-08,
|
| 6414 |
+
"loss": 0.886353874206543,
|
| 6415 |
+
"mean_token_accuracy": 0.7987162381410599,
|
| 6416 |
+
"num_tokens": 13482642.0,
|
| 6417 |
+
"step": 5780
|
| 6418 |
+
},
|
| 6419 |
+
{
|
| 6420 |
+
"entropy": 0.8754376731812954,
|
| 6421 |
+
"epoch": 0.9972871721999742,
|
| 6422 |
+
"grad_norm": 0.3896678388118744,
|
| 6423 |
+
"learning_rate": 5.856011023079573e-08,
|
| 6424 |
+
"loss": 0.8634020805358886,
|
| 6425 |
+
"mean_token_accuracy": 0.803331832587719,
|
| 6426 |
+
"num_tokens": 13615820.0,
|
| 6427 |
+
"step": 5790
|
| 6428 |
+
},
|
| 6429 |
+
{
|
| 6430 |
+
"entropy": 0.891757769882679,
|
| 6431 |
+
"epoch": 0.9990096025491969,
|
| 6432 |
+
"grad_norm": 0.3823208808898926,
|
| 6433 |
+
"learning_rate": 2.4112986565621773e-08,
|
| 6434 |
+
"loss": 0.8841445922851563,
|
| 6435 |
+
"mean_token_accuracy": 0.8023776292800904,
|
| 6436 |
+
"num_tokens": 13750656.0,
|
| 6437 |
+
"step": 5800
|
| 6438 |
+
},
|
| 6439 |
+
{
|
| 6440 |
+
"epoch": 0.9990096025491969,
|
| 6441 |
+
"eval_entropy": 0.8907162497113814,
|
| 6442 |
+
"eval_loss": 0.8861898183822632,
|
| 6443 |
+
"eval_mean_token_accuracy": 0.8006550548659512,
|
| 6444 |
+
"eval_num_tokens": 13750656.0,
|
| 6445 |
+
"eval_runtime": 2196.3123,
|
| 6446 |
+
"eval_samples_per_second": 4.699,
|
| 6447 |
+
"eval_steps_per_second": 1.175,
|
| 6448 |
+
"step": 5800
|
| 6449 |
+
},
|
| 6450 |
+
{
|
| 6451 |
+
"epoch": 1.0,
|
| 6452 |
+
"eval_entropy": 0.8907080711504601,
|
| 6453 |
+
"eval_loss": 0.8861844539642334,
|
| 6454 |
+
"eval_mean_token_accuracy": 0.8006432617602077,
|
| 6455 |
+
"eval_num_tokens": 13829843.0,
|
| 6456 |
+
"eval_runtime": 2194.6346,
|
| 6457 |
+
"eval_samples_per_second": 4.703,
|
| 6458 |
+
"eval_steps_per_second": 1.176,
|
| 6459 |
+
"step": 5806
|
| 6460 |
}
|
| 6461 |
],
|
| 6462 |
"logging_steps": 10,
|
|
|
|
| 6471 |
"should_evaluate": false,
|
| 6472 |
"should_log": false,
|
| 6473 |
"should_save": true,
|
| 6474 |
+
"should_training_stop": true
|
| 6475 |
},
|
| 6476 |
"attributes": {}
|
| 6477 |
}
|
| 6478 |
},
|
| 6479 |
+
"total_flos": 9.53462095817474e+17,
|
| 6480 |
"train_batch_size": 2,
|
| 6481 |
"trial_name": null,
|
| 6482 |
"trial_params": null
|