Training in progress, step 4800, 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 +1335 -3
last-checkpoint/adapter_config.json
CHANGED
|
@@ -31,10 +31,10 @@
|
|
| 31 |
"rank_pattern": {},
|
| 32 |
"revision": null,
|
| 33 |
"target_modules": [
|
| 34 |
-
"v_proj",
|
| 35 |
-
"k_proj",
|
| 36 |
"q_proj",
|
| 37 |
-
"
|
|
|
|
|
|
|
| 38 |
],
|
| 39 |
"target_parameters": null,
|
| 40 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 31 |
"rank_pattern": {},
|
| 32 |
"revision": null,
|
| 33 |
"target_modules": [
|
|
|
|
|
|
|
| 34 |
"q_proj",
|
| 35 |
+
"v_proj",
|
| 36 |
+
"o_proj",
|
| 37 |
+
"k_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 sha256:52217ccaabbb2c9a520ff6c890c15d98aa47a4c28fad51432c095d328a2a29be
|
| 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:a6af5e95ef3e872afad1bdf1fdf3ea5f5155675c6586239ad660ccd39c37347e
|
| 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:a82aeb7d859423fae1de432598feb0f08060628848416302157961cd81d904dc
|
| 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 sha256:51933555f7d7492da069489fcbf900fea3e97204e288093912746efcc0733e6e
|
| 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 sha256:7d223d319f4b8402bb6d092355111dac11a4bb47d6e23f14a8f4c5bc3c5b9c8f
|
| 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:dd465d73b04f6c59912627375bc7905b039595a268513d7eb113da343a1d81e2
|
| 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": 0.
|
| 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,
|
|
@@ -4004,6 +4004,1338 @@
|
|
| 4004 |
"eval_samples_per_second": 5.088,
|
| 4005 |
"eval_steps_per_second": 1.272,
|
| 4006 |
"step": 3600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4007 |
}
|
| 4008 |
],
|
| 4009 |
"logging_steps": 10,
|
|
@@ -4023,7 +5355,7 @@
|
|
| 4023 |
"attributes": {}
|
| 4024 |
}
|
| 4025 |
},
|
| 4026 |
-
"total_flos":
|
| 4027 |
"train_batch_size": 2,
|
| 4028 |
"trial_name": null,
|
| 4029 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.8267665676269216,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 4800,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 4004 |
"eval_samples_per_second": 5.088,
|
| 4005 |
"eval_steps_per_second": 1.272,
|
| 4006 |
"step": 3600
|
| 4007 |
+
},
|
| 4008 |
+
{
|
| 4009 |
+
"entropy": 0.8791561007499695,
|
| 4010 |
+
"epoch": 0.621797356069414,
|
| 4011 |
+
"grad_norm": 0.4086858332157135,
|
| 4012 |
+
"learning_rate": 7.568033069238719e-06,
|
| 4013 |
+
"loss": 0.8722482681274414,
|
| 4014 |
+
"mean_token_accuracy": 0.8025152161717415,
|
| 4015 |
+
"num_tokens": 132977.0,
|
| 4016 |
+
"step": 3610
|
| 4017 |
+
},
|
| 4018 |
+
{
|
| 4019 |
+
"entropy": 0.899948114156723,
|
| 4020 |
+
"epoch": 0.6235197864186367,
|
| 4021 |
+
"grad_norm": 0.46601250767707825,
|
| 4022 |
+
"learning_rate": 7.533585945573545e-06,
|
| 4023 |
+
"loss": 0.8855493545532227,
|
| 4024 |
+
"mean_token_accuracy": 0.800511172413826,
|
| 4025 |
+
"num_tokens": 272879.0,
|
| 4026 |
+
"step": 3620
|
| 4027 |
+
},
|
| 4028 |
+
{
|
| 4029 |
+
"entropy": 0.8673037841916085,
|
| 4030 |
+
"epoch": 0.6252422167678594,
|
| 4031 |
+
"grad_norm": 0.4145843982696533,
|
| 4032 |
+
"learning_rate": 7.499138821908372e-06,
|
| 4033 |
+
"loss": 0.8640913963317871,
|
| 4034 |
+
"mean_token_accuracy": 0.803620132803917,
|
| 4035 |
+
"num_tokens": 410749.0,
|
| 4036 |
+
"step": 3630
|
| 4037 |
+
},
|
| 4038 |
+
{
|
| 4039 |
+
"entropy": 0.8860716938972473,
|
| 4040 |
+
"epoch": 0.6269646471170822,
|
| 4041 |
+
"grad_norm": 0.42249223589897156,
|
| 4042 |
+
"learning_rate": 7.464691698243198e-06,
|
| 4043 |
+
"loss": 0.8842039108276367,
|
| 4044 |
+
"mean_token_accuracy": 0.7986636623740196,
|
| 4045 |
+
"num_tokens": 547722.0,
|
| 4046 |
+
"step": 3640
|
| 4047 |
+
},
|
| 4048 |
+
{
|
| 4049 |
+
"entropy": 0.8954877465963363,
|
| 4050 |
+
"epoch": 0.6286870774663049,
|
| 4051 |
+
"grad_norm": 0.3925943672657013,
|
| 4052 |
+
"learning_rate": 7.4302445745780235e-06,
|
| 4053 |
+
"loss": 0.8855221748352051,
|
| 4054 |
+
"mean_token_accuracy": 0.7995046123862266,
|
| 4055 |
+
"num_tokens": 684572.0,
|
| 4056 |
+
"step": 3650
|
| 4057 |
+
},
|
| 4058 |
+
{
|
| 4059 |
+
"entropy": 0.8849117711186409,
|
| 4060 |
+
"epoch": 0.6304095078155277,
|
| 4061 |
+
"grad_norm": 0.4629289507865906,
|
| 4062 |
+
"learning_rate": 7.395797450912849e-06,
|
| 4063 |
+
"loss": 0.8729182243347168,
|
| 4064 |
+
"mean_token_accuracy": 0.8043904855847359,
|
| 4065 |
+
"num_tokens": 827644.0,
|
| 4066 |
+
"step": 3660
|
| 4067 |
+
},
|
| 4068 |
+
{
|
| 4069 |
+
"entropy": 0.8804656386375427,
|
| 4070 |
+
"epoch": 0.6321319381647504,
|
| 4071 |
+
"grad_norm": 0.46771207451820374,
|
| 4072 |
+
"learning_rate": 7.361350327247675e-06,
|
| 4073 |
+
"loss": 0.8755840301513672,
|
| 4074 |
+
"mean_token_accuracy": 0.8031970843672752,
|
| 4075 |
+
"num_tokens": 962324.0,
|
| 4076 |
+
"step": 3670
|
| 4077 |
+
},
|
| 4078 |
+
{
|
| 4079 |
+
"entropy": 0.9434955894947052,
|
| 4080 |
+
"epoch": 0.6338543685139733,
|
| 4081 |
+
"grad_norm": 0.5211445689201355,
|
| 4082 |
+
"learning_rate": 7.326903203582502e-06,
|
| 4083 |
+
"loss": 0.9384965896606445,
|
| 4084 |
+
"mean_token_accuracy": 0.7892407014966011,
|
| 4085 |
+
"num_tokens": 1098423.0,
|
| 4086 |
+
"step": 3680
|
| 4087 |
+
},
|
| 4088 |
+
{
|
| 4089 |
+
"entropy": 0.8946269884705543,
|
| 4090 |
+
"epoch": 0.635576798863196,
|
| 4091 |
+
"grad_norm": 0.5688809752464294,
|
| 4092 |
+
"learning_rate": 7.292456079917327e-06,
|
| 4093 |
+
"loss": 0.8830915451049804,
|
| 4094 |
+
"mean_token_accuracy": 0.8004635199904442,
|
| 4095 |
+
"num_tokens": 1236560.0,
|
| 4096 |
+
"step": 3690
|
| 4097 |
+
},
|
| 4098 |
+
{
|
| 4099 |
+
"entropy": 0.8900304272770881,
|
| 4100 |
+
"epoch": 0.6372992292124188,
|
| 4101 |
+
"grad_norm": 0.38199931383132935,
|
| 4102 |
+
"learning_rate": 7.258008956252154e-06,
|
| 4103 |
+
"loss": 0.8838729858398438,
|
| 4104 |
+
"mean_token_accuracy": 0.8015081480145454,
|
| 4105 |
+
"num_tokens": 1374095.0,
|
| 4106 |
+
"step": 3700
|
| 4107 |
+
},
|
| 4108 |
+
{
|
| 4109 |
+
"epoch": 0.6372992292124188,
|
| 4110 |
+
"eval_entropy": 0.8987031751527014,
|
| 4111 |
+
"eval_loss": 0.8895877599716187,
|
| 4112 |
+
"eval_mean_token_accuracy": 0.7999683152285254,
|
| 4113 |
+
"eval_num_tokens": 1374095.0,
|
| 4114 |
+
"eval_runtime": 1996.9933,
|
| 4115 |
+
"eval_samples_per_second": 5.168,
|
| 4116 |
+
"eval_steps_per_second": 1.292,
|
| 4117 |
+
"step": 3700
|
| 4118 |
+
},
|
| 4119 |
+
{
|
| 4120 |
+
"entropy": 0.8338712751865387,
|
| 4121 |
+
"epoch": 0.6390216595616415,
|
| 4122 |
+
"grad_norm": 0.38810011744499207,
|
| 4123 |
+
"learning_rate": 7.22356183258698e-06,
|
| 4124 |
+
"loss": 0.8263821601867676,
|
| 4125 |
+
"mean_token_accuracy": 0.8129131481051445,
|
| 4126 |
+
"num_tokens": 1513665.0,
|
| 4127 |
+
"step": 3710
|
| 4128 |
+
},
|
| 4129 |
+
{
|
| 4130 |
+
"entropy": 0.8491598337888717,
|
| 4131 |
+
"epoch": 0.6407440899108642,
|
| 4132 |
+
"grad_norm": 0.5338125228881836,
|
| 4133 |
+
"learning_rate": 7.189114708921806e-06,
|
| 4134 |
+
"loss": 0.8426370620727539,
|
| 4135 |
+
"mean_token_accuracy": 0.8097073540091515,
|
| 4136 |
+
"num_tokens": 1655522.0,
|
| 4137 |
+
"step": 3720
|
| 4138 |
+
},
|
| 4139 |
+
{
|
| 4140 |
+
"entropy": 0.8653292268514633,
|
| 4141 |
+
"epoch": 0.642466520260087,
|
| 4142 |
+
"grad_norm": 0.4272521734237671,
|
| 4143 |
+
"learning_rate": 7.154667585256631e-06,
|
| 4144 |
+
"loss": 0.8501546859741211,
|
| 4145 |
+
"mean_token_accuracy": 0.8075862035155297,
|
| 4146 |
+
"num_tokens": 1792787.0,
|
| 4147 |
+
"step": 3730
|
| 4148 |
+
},
|
| 4149 |
+
{
|
| 4150 |
+
"entropy": 0.86524413600564,
|
| 4151 |
+
"epoch": 0.6441889506093097,
|
| 4152 |
+
"grad_norm": 0.40841034054756165,
|
| 4153 |
+
"learning_rate": 7.120220461591457e-06,
|
| 4154 |
+
"loss": 0.8654015541076661,
|
| 4155 |
+
"mean_token_accuracy": 0.8041766241192818,
|
| 4156 |
+
"num_tokens": 1931208.0,
|
| 4157 |
+
"step": 3740
|
| 4158 |
+
},
|
| 4159 |
+
{
|
| 4160 |
+
"entropy": 0.8816385686397552,
|
| 4161 |
+
"epoch": 0.6459113809585325,
|
| 4162 |
+
"grad_norm": 0.4200495183467865,
|
| 4163 |
+
"learning_rate": 7.085773337926284e-06,
|
| 4164 |
+
"loss": 0.8681506156921387,
|
| 4165 |
+
"mean_token_accuracy": 0.8023644983768463,
|
| 4166 |
+
"num_tokens": 2069218.0,
|
| 4167 |
+
"step": 3750
|
| 4168 |
+
},
|
| 4169 |
+
{
|
| 4170 |
+
"entropy": 0.8531557321548462,
|
| 4171 |
+
"epoch": 0.6476338113077552,
|
| 4172 |
+
"grad_norm": 0.4340897500514984,
|
| 4173 |
+
"learning_rate": 7.051326214261109e-06,
|
| 4174 |
+
"loss": 0.8407665252685547,
|
| 4175 |
+
"mean_token_accuracy": 0.8092893213033676,
|
| 4176 |
+
"num_tokens": 2207220.0,
|
| 4177 |
+
"step": 3760
|
| 4178 |
+
},
|
| 4179 |
+
{
|
| 4180 |
+
"entropy": 0.9071142002940178,
|
| 4181 |
+
"epoch": 0.649356241656978,
|
| 4182 |
+
"grad_norm": 0.4704848527908325,
|
| 4183 |
+
"learning_rate": 7.016879090595936e-06,
|
| 4184 |
+
"loss": 0.9017156600952149,
|
| 4185 |
+
"mean_token_accuracy": 0.7971421957015992,
|
| 4186 |
+
"num_tokens": 2341438.0,
|
| 4187 |
+
"step": 3770
|
| 4188 |
+
},
|
| 4189 |
+
{
|
| 4190 |
+
"entropy": 0.9055481895804405,
|
| 4191 |
+
"epoch": 0.6510786720062007,
|
| 4192 |
+
"grad_norm": 0.39707693457603455,
|
| 4193 |
+
"learning_rate": 6.982431966930762e-06,
|
| 4194 |
+
"loss": 0.9052577972412109,
|
| 4195 |
+
"mean_token_accuracy": 0.7952045202255249,
|
| 4196 |
+
"num_tokens": 2482296.0,
|
| 4197 |
+
"step": 3780
|
| 4198 |
+
},
|
| 4199 |
+
{
|
| 4200 |
+
"entropy": 0.8833331823348999,
|
| 4201 |
+
"epoch": 0.6528011023554235,
|
| 4202 |
+
"grad_norm": 0.416567862033844,
|
| 4203 |
+
"learning_rate": 6.9479848432655884e-06,
|
| 4204 |
+
"loss": 0.8772726058959961,
|
| 4205 |
+
"mean_token_accuracy": 0.8013244524598122,
|
| 4206 |
+
"num_tokens": 2622763.0,
|
| 4207 |
+
"step": 3790
|
| 4208 |
+
},
|
| 4209 |
+
{
|
| 4210 |
+
"entropy": 0.8962781682610512,
|
| 4211 |
+
"epoch": 0.6545235327046462,
|
| 4212 |
+
"grad_norm": 0.41791078448295593,
|
| 4213 |
+
"learning_rate": 6.9135377196004135e-06,
|
| 4214 |
+
"loss": 0.8844639778137207,
|
| 4215 |
+
"mean_token_accuracy": 0.8016847550868988,
|
| 4216 |
+
"num_tokens": 2756826.0,
|
| 4217 |
+
"step": 3800
|
| 4218 |
+
},
|
| 4219 |
+
{
|
| 4220 |
+
"epoch": 0.6545235327046462,
|
| 4221 |
+
"eval_entropy": 0.8984109290876577,
|
| 4222 |
+
"eval_loss": 0.8893030881881714,
|
| 4223 |
+
"eval_mean_token_accuracy": 0.8000404531297088,
|
| 4224 |
+
"eval_num_tokens": 2756826.0,
|
| 4225 |
+
"eval_runtime": 1997.5778,
|
| 4226 |
+
"eval_samples_per_second": 5.167,
|
| 4227 |
+
"eval_steps_per_second": 1.292,
|
| 4228 |
+
"step": 3800
|
| 4229 |
+
},
|
| 4230 |
+
{
|
| 4231 |
+
"entropy": 0.8708895847201348,
|
| 4232 |
+
"epoch": 0.6562459630538691,
|
| 4233 |
+
"grad_norm": 0.3722359538078308,
|
| 4234 |
+
"learning_rate": 6.879090595935239e-06,
|
| 4235 |
+
"loss": 0.8560014724731445,
|
| 4236 |
+
"mean_token_accuracy": 0.8055260598659515,
|
| 4237 |
+
"num_tokens": 2896570.0,
|
| 4238 |
+
"step": 3810
|
| 4239 |
+
},
|
| 4240 |
+
{
|
| 4241 |
+
"entropy": 0.8795327812433242,
|
| 4242 |
+
"epoch": 0.6579683934030918,
|
| 4243 |
+
"grad_norm": 0.4038887917995453,
|
| 4244 |
+
"learning_rate": 6.844643472270066e-06,
|
| 4245 |
+
"loss": 0.8748892784118653,
|
| 4246 |
+
"mean_token_accuracy": 0.8010203137993812,
|
| 4247 |
+
"num_tokens": 3036528.0,
|
| 4248 |
+
"step": 3820
|
| 4249 |
+
},
|
| 4250 |
+
{
|
| 4251 |
+
"entropy": 0.8723071977496147,
|
| 4252 |
+
"epoch": 0.6596908237523145,
|
| 4253 |
+
"grad_norm": 0.38578343391418457,
|
| 4254 |
+
"learning_rate": 6.810196348604892e-06,
|
| 4255 |
+
"loss": 0.8659199714660645,
|
| 4256 |
+
"mean_token_accuracy": 0.8032226487994194,
|
| 4257 |
+
"num_tokens": 3173432.0,
|
| 4258 |
+
"step": 3830
|
| 4259 |
+
},
|
| 4260 |
+
{
|
| 4261 |
+
"entropy": 0.8979651927947998,
|
| 4262 |
+
"epoch": 0.6614132541015373,
|
| 4263 |
+
"grad_norm": 0.4457395672798157,
|
| 4264 |
+
"learning_rate": 6.775749224939719e-06,
|
| 4265 |
+
"loss": 0.8884675979614258,
|
| 4266 |
+
"mean_token_accuracy": 0.8004711583256722,
|
| 4267 |
+
"num_tokens": 3309359.0,
|
| 4268 |
+
"step": 3840
|
| 4269 |
+
},
|
| 4270 |
+
{
|
| 4271 |
+
"entropy": 0.8551001802086831,
|
| 4272 |
+
"epoch": 0.66313568445076,
|
| 4273 |
+
"grad_norm": 0.4345524311065674,
|
| 4274 |
+
"learning_rate": 6.741302101274544e-06,
|
| 4275 |
+
"loss": 0.8385509490966797,
|
| 4276 |
+
"mean_token_accuracy": 0.8087713450193406,
|
| 4277 |
+
"num_tokens": 3444443.0,
|
| 4278 |
+
"step": 3850
|
| 4279 |
+
},
|
| 4280 |
+
{
|
| 4281 |
+
"entropy": 0.891946230828762,
|
| 4282 |
+
"epoch": 0.6648581147999828,
|
| 4283 |
+
"grad_norm": 0.4427470862865448,
|
| 4284 |
+
"learning_rate": 6.7068549776093705e-06,
|
| 4285 |
+
"loss": 0.8853155136108398,
|
| 4286 |
+
"mean_token_accuracy": 0.7998019322752953,
|
| 4287 |
+
"num_tokens": 3582885.0,
|
| 4288 |
+
"step": 3860
|
| 4289 |
+
},
|
| 4290 |
+
{
|
| 4291 |
+
"entropy": 0.8642865002155304,
|
| 4292 |
+
"epoch": 0.6665805451492055,
|
| 4293 |
+
"grad_norm": 0.4064953327178955,
|
| 4294 |
+
"learning_rate": 6.672407853944196e-06,
|
| 4295 |
+
"loss": 0.8612741470336914,
|
| 4296 |
+
"mean_token_accuracy": 0.802153280377388,
|
| 4297 |
+
"num_tokens": 3722762.0,
|
| 4298 |
+
"step": 3870
|
| 4299 |
+
},
|
| 4300 |
+
{
|
| 4301 |
+
"entropy": 0.8654727056622505,
|
| 4302 |
+
"epoch": 0.6683029754984283,
|
| 4303 |
+
"grad_norm": 0.44759929180145264,
|
| 4304 |
+
"learning_rate": 6.637960730279023e-06,
|
| 4305 |
+
"loss": 0.8589719772338867,
|
| 4306 |
+
"mean_token_accuracy": 0.8055579513311386,
|
| 4307 |
+
"num_tokens": 3862098.0,
|
| 4308 |
+
"step": 3880
|
| 4309 |
+
},
|
| 4310 |
+
{
|
| 4311 |
+
"entropy": 0.9094881787896156,
|
| 4312 |
+
"epoch": 0.670025405847651,
|
| 4313 |
+
"grad_norm": 0.41999101638793945,
|
| 4314 |
+
"learning_rate": 6.603513606613848e-06,
|
| 4315 |
+
"loss": 0.901517391204834,
|
| 4316 |
+
"mean_token_accuracy": 0.798083683848381,
|
| 4317 |
+
"num_tokens": 4000549.0,
|
| 4318 |
+
"step": 3890
|
| 4319 |
+
},
|
| 4320 |
+
{
|
| 4321 |
+
"entropy": 0.9040753439068794,
|
| 4322 |
+
"epoch": 0.6717478361968738,
|
| 4323 |
+
"grad_norm": 0.39481016993522644,
|
| 4324 |
+
"learning_rate": 6.569066482948674e-06,
|
| 4325 |
+
"loss": 0.8963336944580078,
|
| 4326 |
+
"mean_token_accuracy": 0.7986732751131058,
|
| 4327 |
+
"num_tokens": 4137043.0,
|
| 4328 |
+
"step": 3900
|
| 4329 |
+
},
|
| 4330 |
+
{
|
| 4331 |
+
"epoch": 0.6717478361968738,
|
| 4332 |
+
"eval_entropy": 0.8978190375062579,
|
| 4333 |
+
"eval_loss": 0.8890023827552795,
|
| 4334 |
+
"eval_mean_token_accuracy": 0.8000849456530308,
|
| 4335 |
+
"eval_num_tokens": 4137043.0,
|
| 4336 |
+
"eval_runtime": 1997.0726,
|
| 4337 |
+
"eval_samples_per_second": 5.168,
|
| 4338 |
+
"eval_steps_per_second": 1.292,
|
| 4339 |
+
"step": 3900
|
| 4340 |
+
},
|
| 4341 |
+
{
|
| 4342 |
+
"entropy": 0.888623583316803,
|
| 4343 |
+
"epoch": 0.6734702665460965,
|
| 4344 |
+
"grad_norm": 0.3936791718006134,
|
| 4345 |
+
"learning_rate": 6.534619359283501e-06,
|
| 4346 |
+
"loss": 0.8851858139038086,
|
| 4347 |
+
"mean_token_accuracy": 0.80007184445858,
|
| 4348 |
+
"num_tokens": 4273008.0,
|
| 4349 |
+
"step": 3910
|
| 4350 |
+
},
|
| 4351 |
+
{
|
| 4352 |
+
"entropy": 0.9054401487112045,
|
| 4353 |
+
"epoch": 0.6751926968953192,
|
| 4354 |
+
"grad_norm": 0.4450265169143677,
|
| 4355 |
+
"learning_rate": 6.500172235618326e-06,
|
| 4356 |
+
"loss": 0.8935989379882813,
|
| 4357 |
+
"mean_token_accuracy": 0.7988793134689331,
|
| 4358 |
+
"num_tokens": 4415036.0,
|
| 4359 |
+
"step": 3920
|
| 4360 |
+
},
|
| 4361 |
+
{
|
| 4362 |
+
"entropy": 0.8839291751384735,
|
| 4363 |
+
"epoch": 0.6769151272445421,
|
| 4364 |
+
"grad_norm": 0.44173672795295715,
|
| 4365 |
+
"learning_rate": 6.4657251119531525e-06,
|
| 4366 |
+
"loss": 0.8775888442993164,
|
| 4367 |
+
"mean_token_accuracy": 0.8007264167070389,
|
| 4368 |
+
"num_tokens": 4552708.0,
|
| 4369 |
+
"step": 3930
|
| 4370 |
+
},
|
| 4371 |
+
{
|
| 4372 |
+
"entropy": 0.8589925251901149,
|
| 4373 |
+
"epoch": 0.6786375575937648,
|
| 4374 |
+
"grad_norm": 0.3620625436306,
|
| 4375 |
+
"learning_rate": 6.4312779882879784e-06,
|
| 4376 |
+
"loss": 0.8551578521728516,
|
| 4377 |
+
"mean_token_accuracy": 0.8064107894897461,
|
| 4378 |
+
"num_tokens": 4696290.0,
|
| 4379 |
+
"step": 3940
|
| 4380 |
+
},
|
| 4381 |
+
{
|
| 4382 |
+
"entropy": 0.8445931300520897,
|
| 4383 |
+
"epoch": 0.6803599879429876,
|
| 4384 |
+
"grad_norm": 0.4519362449645996,
|
| 4385 |
+
"learning_rate": 6.396830864622805e-06,
|
| 4386 |
+
"loss": 0.8345125198364258,
|
| 4387 |
+
"mean_token_accuracy": 0.8092260167002678,
|
| 4388 |
+
"num_tokens": 4832087.0,
|
| 4389 |
+
"step": 3950
|
| 4390 |
+
},
|
| 4391 |
+
{
|
| 4392 |
+
"entropy": 0.9084868378937244,
|
| 4393 |
+
"epoch": 0.6820824182922103,
|
| 4394 |
+
"grad_norm": 0.40370234847068787,
|
| 4395 |
+
"learning_rate": 6.36238374095763e-06,
|
| 4396 |
+
"loss": 0.901630973815918,
|
| 4397 |
+
"mean_token_accuracy": 0.7952028334140777,
|
| 4398 |
+
"num_tokens": 4972261.0,
|
| 4399 |
+
"step": 3960
|
| 4400 |
+
},
|
| 4401 |
+
{
|
| 4402 |
+
"entropy": 0.8734813243150711,
|
| 4403 |
+
"epoch": 0.6838048486414331,
|
| 4404 |
+
"grad_norm": 0.4129689335823059,
|
| 4405 |
+
"learning_rate": 6.327936617292456e-06,
|
| 4406 |
+
"loss": 0.8525594711303711,
|
| 4407 |
+
"mean_token_accuracy": 0.806535467505455,
|
| 4408 |
+
"num_tokens": 5105943.0,
|
| 4409 |
+
"step": 3970
|
| 4410 |
+
},
|
| 4411 |
+
{
|
| 4412 |
+
"entropy": 0.8367196694016457,
|
| 4413 |
+
"epoch": 0.6855272789906558,
|
| 4414 |
+
"grad_norm": 0.38863131403923035,
|
| 4415 |
+
"learning_rate": 6.293489493627283e-06,
|
| 4416 |
+
"loss": 0.8323159217834473,
|
| 4417 |
+
"mean_token_accuracy": 0.8115260645747184,
|
| 4418 |
+
"num_tokens": 5241657.0,
|
| 4419 |
+
"step": 3980
|
| 4420 |
+
},
|
| 4421 |
+
{
|
| 4422 |
+
"entropy": 0.8844600863754749,
|
| 4423 |
+
"epoch": 0.6872497093398786,
|
| 4424 |
+
"grad_norm": 0.4728356599807739,
|
| 4425 |
+
"learning_rate": 6.259042369962109e-06,
|
| 4426 |
+
"loss": 0.8816679000854493,
|
| 4427 |
+
"mean_token_accuracy": 0.800590705871582,
|
| 4428 |
+
"num_tokens": 5381973.0,
|
| 4429 |
+
"step": 3990
|
| 4430 |
+
},
|
| 4431 |
+
{
|
| 4432 |
+
"entropy": 0.865593820810318,
|
| 4433 |
+
"epoch": 0.6889721396891013,
|
| 4434 |
+
"grad_norm": 0.3811900317668915,
|
| 4435 |
+
"learning_rate": 6.224595246296935e-06,
|
| 4436 |
+
"loss": 0.8632452011108398,
|
| 4437 |
+
"mean_token_accuracy": 0.8051433145999909,
|
| 4438 |
+
"num_tokens": 5520841.0,
|
| 4439 |
+
"step": 4000
|
| 4440 |
+
},
|
| 4441 |
+
{
|
| 4442 |
+
"epoch": 0.6889721396891013,
|
| 4443 |
+
"eval_entropy": 0.8943157961367637,
|
| 4444 |
+
"eval_loss": 0.8887593150138855,
|
| 4445 |
+
"eval_mean_token_accuracy": 0.8001742187002656,
|
| 4446 |
+
"eval_num_tokens": 5520841.0,
|
| 4447 |
+
"eval_runtime": 1998.5827,
|
| 4448 |
+
"eval_samples_per_second": 5.164,
|
| 4449 |
+
"eval_steps_per_second": 1.291,
|
| 4450 |
+
"step": 4000
|
| 4451 |
+
},
|
| 4452 |
+
{
|
| 4453 |
+
"entropy": 0.8643890962004661,
|
| 4454 |
+
"epoch": 0.6906945700383241,
|
| 4455 |
+
"grad_norm": 0.3872925937175751,
|
| 4456 |
+
"learning_rate": 6.1901481226317605e-06,
|
| 4457 |
+
"loss": 0.8602989196777344,
|
| 4458 |
+
"mean_token_accuracy": 0.8069830879569053,
|
| 4459 |
+
"num_tokens": 5658173.0,
|
| 4460 |
+
"step": 4010
|
| 4461 |
+
},
|
| 4462 |
+
{
|
| 4463 |
+
"entropy": 0.8567002490162849,
|
| 4464 |
+
"epoch": 0.6924170003875468,
|
| 4465 |
+
"grad_norm": 0.35181790590286255,
|
| 4466 |
+
"learning_rate": 6.155700998966587e-06,
|
| 4467 |
+
"loss": 0.8468066215515136,
|
| 4468 |
+
"mean_token_accuracy": 0.8063052579760551,
|
| 4469 |
+
"num_tokens": 5799573.0,
|
| 4470 |
+
"step": 4020
|
| 4471 |
+
},
|
| 4472 |
+
{
|
| 4473 |
+
"entropy": 0.8518275752663612,
|
| 4474 |
+
"epoch": 0.6941394307367695,
|
| 4475 |
+
"grad_norm": 0.39484772086143494,
|
| 4476 |
+
"learning_rate": 6.121253875301413e-06,
|
| 4477 |
+
"loss": 0.8427839279174805,
|
| 4478 |
+
"mean_token_accuracy": 0.8105209484696388,
|
| 4479 |
+
"num_tokens": 5935743.0,
|
| 4480 |
+
"step": 4030
|
| 4481 |
+
},
|
| 4482 |
+
{
|
| 4483 |
+
"entropy": 0.8372007101774216,
|
| 4484 |
+
"epoch": 0.6958618610859924,
|
| 4485 |
+
"grad_norm": 0.38959771394729614,
|
| 4486 |
+
"learning_rate": 6.086806751636238e-06,
|
| 4487 |
+
"loss": 0.8350957870483399,
|
| 4488 |
+
"mean_token_accuracy": 0.8103716000914574,
|
| 4489 |
+
"num_tokens": 6074734.0,
|
| 4490 |
+
"step": 4040
|
| 4491 |
+
},
|
| 4492 |
+
{
|
| 4493 |
+
"entropy": 0.8757128149271012,
|
| 4494 |
+
"epoch": 0.697584291435215,
|
| 4495 |
+
"grad_norm": 0.4061984717845917,
|
| 4496 |
+
"learning_rate": 6.052359627971065e-06,
|
| 4497 |
+
"loss": 0.8627335548400878,
|
| 4498 |
+
"mean_token_accuracy": 0.8039230793714524,
|
| 4499 |
+
"num_tokens": 6213263.0,
|
| 4500 |
+
"step": 4050
|
| 4501 |
+
},
|
| 4502 |
+
{
|
| 4503 |
+
"entropy": 0.8416135057806968,
|
| 4504 |
+
"epoch": 0.6993067217844379,
|
| 4505 |
+
"grad_norm": 0.4353971779346466,
|
| 4506 |
+
"learning_rate": 6.017912504305891e-06,
|
| 4507 |
+
"loss": 0.8411669731140137,
|
| 4508 |
+
"mean_token_accuracy": 0.8084114342927933,
|
| 4509 |
+
"num_tokens": 6353905.0,
|
| 4510 |
+
"step": 4060
|
| 4511 |
+
},
|
| 4512 |
+
{
|
| 4513 |
+
"entropy": 0.8721702679991722,
|
| 4514 |
+
"epoch": 0.7010291521336606,
|
| 4515 |
+
"grad_norm": 0.40984073281288147,
|
| 4516 |
+
"learning_rate": 5.9834653806407175e-06,
|
| 4517 |
+
"loss": 0.8583096504211426,
|
| 4518 |
+
"mean_token_accuracy": 0.8049539163708687,
|
| 4519 |
+
"num_tokens": 6492756.0,
|
| 4520 |
+
"step": 4070
|
| 4521 |
+
},
|
| 4522 |
+
{
|
| 4523 |
+
"entropy": 0.8773854672908783,
|
| 4524 |
+
"epoch": 0.7027515824828834,
|
| 4525 |
+
"grad_norm": 0.45472994446754456,
|
| 4526 |
+
"learning_rate": 5.9490182569755425e-06,
|
| 4527 |
+
"loss": 0.8659296989440918,
|
| 4528 |
+
"mean_token_accuracy": 0.8020806655287742,
|
| 4529 |
+
"num_tokens": 6628141.0,
|
| 4530 |
+
"step": 4080
|
| 4531 |
+
},
|
| 4532 |
+
{
|
| 4533 |
+
"entropy": 0.8715768173336983,
|
| 4534 |
+
"epoch": 0.7044740128321061,
|
| 4535 |
+
"grad_norm": 0.5028422474861145,
|
| 4536 |
+
"learning_rate": 5.914571133310369e-06,
|
| 4537 |
+
"loss": 0.8667140960693359,
|
| 4538 |
+
"mean_token_accuracy": 0.8024423226714135,
|
| 4539 |
+
"num_tokens": 6767454.0,
|
| 4540 |
+
"step": 4090
|
| 4541 |
+
},
|
| 4542 |
+
{
|
| 4543 |
+
"entropy": 0.7750305764377117,
|
| 4544 |
+
"epoch": 0.7061964431813289,
|
| 4545 |
+
"grad_norm": 0.3967188000679016,
|
| 4546 |
+
"learning_rate": 5.880124009645195e-06,
|
| 4547 |
+
"loss": 0.77204270362854,
|
| 4548 |
+
"mean_token_accuracy": 0.8227861791849136,
|
| 4549 |
+
"num_tokens": 6910506.0,
|
| 4550 |
+
"step": 4100
|
| 4551 |
+
},
|
| 4552 |
+
{
|
| 4553 |
+
"epoch": 0.7061964431813289,
|
| 4554 |
+
"eval_entropy": 0.8902723185316808,
|
| 4555 |
+
"eval_loss": 0.8885339498519897,
|
| 4556 |
+
"eval_mean_token_accuracy": 0.8002320625023618,
|
| 4557 |
+
"eval_num_tokens": 6910506.0,
|
| 4558 |
+
"eval_runtime": 1997.4866,
|
| 4559 |
+
"eval_samples_per_second": 5.167,
|
| 4560 |
+
"eval_steps_per_second": 1.292,
|
| 4561 |
+
"step": 4100
|
| 4562 |
+
},
|
| 4563 |
+
{
|
| 4564 |
+
"entropy": 0.8898505240678787,
|
| 4565 |
+
"epoch": 0.7079188735305516,
|
| 4566 |
+
"grad_norm": 0.3867063820362091,
|
| 4567 |
+
"learning_rate": 5.845676885980022e-06,
|
| 4568 |
+
"loss": 0.8828920364379883,
|
| 4569 |
+
"mean_token_accuracy": 0.8001085966825485,
|
| 4570 |
+
"num_tokens": 7054312.0,
|
| 4571 |
+
"step": 4110
|
| 4572 |
+
},
|
| 4573 |
+
{
|
| 4574 |
+
"entropy": 0.8769096612930298,
|
| 4575 |
+
"epoch": 0.7096413038797743,
|
| 4576 |
+
"grad_norm": 0.4864627718925476,
|
| 4577 |
+
"learning_rate": 5.811229762314847e-06,
|
| 4578 |
+
"loss": 0.8730987548828125,
|
| 4579 |
+
"mean_token_accuracy": 0.8028520822525025,
|
| 4580 |
+
"num_tokens": 7191411.0,
|
| 4581 |
+
"step": 4120
|
| 4582 |
+
},
|
| 4583 |
+
{
|
| 4584 |
+
"entropy": 0.8836300104856492,
|
| 4585 |
+
"epoch": 0.7113637342289971,
|
| 4586 |
+
"grad_norm": 0.4547421634197235,
|
| 4587 |
+
"learning_rate": 5.776782638649673e-06,
|
| 4588 |
+
"loss": 0.8823716163635253,
|
| 4589 |
+
"mean_token_accuracy": 0.8020639196038246,
|
| 4590 |
+
"num_tokens": 7327752.0,
|
| 4591 |
+
"step": 4130
|
| 4592 |
+
},
|
| 4593 |
+
{
|
| 4594 |
+
"entropy": 0.8258942127227783,
|
| 4595 |
+
"epoch": 0.7130861645782198,
|
| 4596 |
+
"grad_norm": 0.48732635378837585,
|
| 4597 |
+
"learning_rate": 5.7423355149844995e-06,
|
| 4598 |
+
"loss": 0.8261510848999023,
|
| 4599 |
+
"mean_token_accuracy": 0.8117264598608017,
|
| 4600 |
+
"num_tokens": 7466507.0,
|
| 4601 |
+
"step": 4140
|
| 4602 |
+
},
|
| 4603 |
+
{
|
| 4604 |
+
"entropy": 0.9056715458631516,
|
| 4605 |
+
"epoch": 0.7148085949274426,
|
| 4606 |
+
"grad_norm": 0.4778401553630829,
|
| 4607 |
+
"learning_rate": 5.7078883913193254e-06,
|
| 4608 |
+
"loss": 0.9034542083740235,
|
| 4609 |
+
"mean_token_accuracy": 0.7989412933588028,
|
| 4610 |
+
"num_tokens": 7602478.0,
|
| 4611 |
+
"step": 4150
|
| 4612 |
+
},
|
| 4613 |
+
{
|
| 4614 |
+
"entropy": 0.8790255658328533,
|
| 4615 |
+
"epoch": 0.7165310252766653,
|
| 4616 |
+
"grad_norm": 0.48434892296791077,
|
| 4617 |
+
"learning_rate": 5.673441267654151e-06,
|
| 4618 |
+
"loss": 0.8668739318847656,
|
| 4619 |
+
"mean_token_accuracy": 0.8019752576947212,
|
| 4620 |
+
"num_tokens": 7743051.0,
|
| 4621 |
+
"step": 4160
|
| 4622 |
+
},
|
| 4623 |
+
{
|
| 4624 |
+
"entropy": 0.8454579740762711,
|
| 4625 |
+
"epoch": 0.7182534556258882,
|
| 4626 |
+
"grad_norm": 0.4142749309539795,
|
| 4627 |
+
"learning_rate": 5.638994143988977e-06,
|
| 4628 |
+
"loss": 0.8422212600708008,
|
| 4629 |
+
"mean_token_accuracy": 0.810345932841301,
|
| 4630 |
+
"num_tokens": 7879350.0,
|
| 4631 |
+
"step": 4170
|
| 4632 |
+
},
|
| 4633 |
+
{
|
| 4634 |
+
"entropy": 0.8710669368505478,
|
| 4635 |
+
"epoch": 0.7199758859751109,
|
| 4636 |
+
"grad_norm": 0.4645155072212219,
|
| 4637 |
+
"learning_rate": 5.604547020323804e-06,
|
| 4638 |
+
"loss": 0.8654325485229493,
|
| 4639 |
+
"mean_token_accuracy": 0.8029890432953835,
|
| 4640 |
+
"num_tokens": 8018257.0,
|
| 4641 |
+
"step": 4180
|
| 4642 |
+
},
|
| 4643 |
+
{
|
| 4644 |
+
"entropy": 0.8802241533994675,
|
| 4645 |
+
"epoch": 0.7216983163243337,
|
| 4646 |
+
"grad_norm": 0.42126893997192383,
|
| 4647 |
+
"learning_rate": 5.57009989665863e-06,
|
| 4648 |
+
"loss": 0.8745281219482421,
|
| 4649 |
+
"mean_token_accuracy": 0.8018666654825211,
|
| 4650 |
+
"num_tokens": 8155431.0,
|
| 4651 |
+
"step": 4190
|
| 4652 |
+
},
|
| 4653 |
+
{
|
| 4654 |
+
"entropy": 0.8723957024514675,
|
| 4655 |
+
"epoch": 0.7234207466735564,
|
| 4656 |
+
"grad_norm": 0.4210054576396942,
|
| 4657 |
+
"learning_rate": 5.535652772993455e-06,
|
| 4658 |
+
"loss": 0.8660004615783692,
|
| 4659 |
+
"mean_token_accuracy": 0.8050700455904007,
|
| 4660 |
+
"num_tokens": 8292905.0,
|
| 4661 |
+
"step": 4200
|
| 4662 |
+
},
|
| 4663 |
+
{
|
| 4664 |
+
"epoch": 0.7234207466735564,
|
| 4665 |
+
"eval_entropy": 0.8844605909212298,
|
| 4666 |
+
"eval_loss": 0.8882371783256531,
|
| 4667 |
+
"eval_mean_token_accuracy": 0.8002364763175273,
|
| 4668 |
+
"eval_num_tokens": 8292905.0,
|
| 4669 |
+
"eval_runtime": 1998.0708,
|
| 4670 |
+
"eval_samples_per_second": 5.165,
|
| 4671 |
+
"eval_steps_per_second": 1.292,
|
| 4672 |
+
"step": 4200
|
| 4673 |
+
},
|
| 4674 |
+
{
|
| 4675 |
+
"entropy": 0.8580236986279488,
|
| 4676 |
+
"epoch": 0.7251431770227792,
|
| 4677 |
+
"grad_norm": 0.4309401512145996,
|
| 4678 |
+
"learning_rate": 5.501205649328282e-06,
|
| 4679 |
+
"loss": 0.8568378448486328,
|
| 4680 |
+
"mean_token_accuracy": 0.8047821834683418,
|
| 4681 |
+
"num_tokens": 8429860.0,
|
| 4682 |
+
"step": 4210
|
| 4683 |
+
},
|
| 4684 |
+
{
|
| 4685 |
+
"entropy": 0.8618510410189628,
|
| 4686 |
+
"epoch": 0.7268656073720019,
|
| 4687 |
+
"grad_norm": 0.4365301728248596,
|
| 4688 |
+
"learning_rate": 5.4667585256631075e-06,
|
| 4689 |
+
"loss": 0.8455228805541992,
|
| 4690 |
+
"mean_token_accuracy": 0.8064143791794777,
|
| 4691 |
+
"num_tokens": 8569302.0,
|
| 4692 |
+
"step": 4220
|
| 4693 |
+
},
|
| 4694 |
+
{
|
| 4695 |
+
"entropy": 0.8439206860959529,
|
| 4696 |
+
"epoch": 0.7285880377212246,
|
| 4697 |
+
"grad_norm": 0.4303825795650482,
|
| 4698 |
+
"learning_rate": 5.432311401997934e-06,
|
| 4699 |
+
"loss": 0.844693946838379,
|
| 4700 |
+
"mean_token_accuracy": 0.8102762028574944,
|
| 4701 |
+
"num_tokens": 8708885.0,
|
| 4702 |
+
"step": 4230
|
| 4703 |
+
},
|
| 4704 |
+
{
|
| 4705 |
+
"entropy": 0.903216864168644,
|
| 4706 |
+
"epoch": 0.7303104680704474,
|
| 4707 |
+
"grad_norm": 0.38660794496536255,
|
| 4708 |
+
"learning_rate": 5.397864278332759e-06,
|
| 4709 |
+
"loss": 0.9013440132141113,
|
| 4710 |
+
"mean_token_accuracy": 0.7975545451045036,
|
| 4711 |
+
"num_tokens": 8849052.0,
|
| 4712 |
+
"step": 4240
|
| 4713 |
+
},
|
| 4714 |
+
{
|
| 4715 |
+
"entropy": 0.909390689432621,
|
| 4716 |
+
"epoch": 0.7320328984196701,
|
| 4717 |
+
"grad_norm": 0.39681291580200195,
|
| 4718 |
+
"learning_rate": 5.363417154667586e-06,
|
| 4719 |
+
"loss": 0.9025371551513672,
|
| 4720 |
+
"mean_token_accuracy": 0.7976114839315415,
|
| 4721 |
+
"num_tokens": 8991130.0,
|
| 4722 |
+
"step": 4250
|
| 4723 |
+
},
|
| 4724 |
+
{
|
| 4725 |
+
"entropy": 0.8940975949168205,
|
| 4726 |
+
"epoch": 0.7337553287688929,
|
| 4727 |
+
"grad_norm": 0.4652435779571533,
|
| 4728 |
+
"learning_rate": 5.328970031002412e-06,
|
| 4729 |
+
"loss": 0.8880777359008789,
|
| 4730 |
+
"mean_token_accuracy": 0.796917274594307,
|
| 4731 |
+
"num_tokens": 9129200.0,
|
| 4732 |
+
"step": 4260
|
| 4733 |
+
},
|
| 4734 |
+
{
|
| 4735 |
+
"entropy": 0.9148185849189758,
|
| 4736 |
+
"epoch": 0.7354777591181156,
|
| 4737 |
+
"grad_norm": 0.4554401636123657,
|
| 4738 |
+
"learning_rate": 5.294522907337237e-06,
|
| 4739 |
+
"loss": 0.8993970870971679,
|
| 4740 |
+
"mean_token_accuracy": 0.7963166296482086,
|
| 4741 |
+
"num_tokens": 9272121.0,
|
| 4742 |
+
"step": 4270
|
| 4743 |
+
},
|
| 4744 |
+
{
|
| 4745 |
+
"entropy": 0.8930785924196243,
|
| 4746 |
+
"epoch": 0.7372001894673385,
|
| 4747 |
+
"grad_norm": 0.4533047378063202,
|
| 4748 |
+
"learning_rate": 5.260075783672064e-06,
|
| 4749 |
+
"loss": 0.8878902435302735,
|
| 4750 |
+
"mean_token_accuracy": 0.7992581784725189,
|
| 4751 |
+
"num_tokens": 9412352.0,
|
| 4752 |
+
"step": 4280
|
| 4753 |
+
},
|
| 4754 |
+
{
|
| 4755 |
+
"entropy": 0.8735776335000992,
|
| 4756 |
+
"epoch": 0.7389226198165612,
|
| 4757 |
+
"grad_norm": 0.42845964431762695,
|
| 4758 |
+
"learning_rate": 5.2256286600068895e-06,
|
| 4759 |
+
"loss": 0.8622098922729492,
|
| 4760 |
+
"mean_token_accuracy": 0.8053018420934677,
|
| 4761 |
+
"num_tokens": 9550914.0,
|
| 4762 |
+
"step": 4290
|
| 4763 |
+
},
|
| 4764 |
+
{
|
| 4765 |
+
"entropy": 0.8958984062075614,
|
| 4766 |
+
"epoch": 0.740645050165784,
|
| 4767 |
+
"grad_norm": 0.4025866389274597,
|
| 4768 |
+
"learning_rate": 5.191181536341716e-06,
|
| 4769 |
+
"loss": 0.8895586013793946,
|
| 4770 |
+
"mean_token_accuracy": 0.7969679340720177,
|
| 4771 |
+
"num_tokens": 9685816.0,
|
| 4772 |
+
"step": 4300
|
| 4773 |
+
},
|
| 4774 |
+
{
|
| 4775 |
+
"epoch": 0.740645050165784,
|
| 4776 |
+
"eval_entropy": 0.8858984069092612,
|
| 4777 |
+
"eval_loss": 0.8879902362823486,
|
| 4778 |
+
"eval_mean_token_accuracy": 0.8002524884206202,
|
| 4779 |
+
"eval_num_tokens": 9685816.0,
|
| 4780 |
+
"eval_runtime": 1997.672,
|
| 4781 |
+
"eval_samples_per_second": 5.167,
|
| 4782 |
+
"eval_steps_per_second": 1.292,
|
| 4783 |
+
"step": 4300
|
| 4784 |
+
},
|
| 4785 |
+
{
|
| 4786 |
+
"entropy": 0.8693854689598084,
|
| 4787 |
+
"epoch": 0.7423674805150067,
|
| 4788 |
+
"grad_norm": 0.4462966024875641,
|
| 4789 |
+
"learning_rate": 5.156734412676541e-06,
|
| 4790 |
+
"loss": 0.8671138763427735,
|
| 4791 |
+
"mean_token_accuracy": 0.8038399279117584,
|
| 4792 |
+
"num_tokens": 9822350.0,
|
| 4793 |
+
"step": 4310
|
| 4794 |
+
},
|
| 4795 |
+
{
|
| 4796 |
+
"entropy": 0.8892764195799827,
|
| 4797 |
+
"epoch": 0.7440899108642294,
|
| 4798 |
+
"grad_norm": 0.4320750832557678,
|
| 4799 |
+
"learning_rate": 5.122287289011368e-06,
|
| 4800 |
+
"loss": 0.8706260681152344,
|
| 4801 |
+
"mean_token_accuracy": 0.8018735349178314,
|
| 4802 |
+
"num_tokens": 9958783.0,
|
| 4803 |
+
"step": 4320
|
| 4804 |
+
},
|
| 4805 |
+
{
|
| 4806 |
+
"entropy": 0.8506387054920197,
|
| 4807 |
+
"epoch": 0.7458123412134522,
|
| 4808 |
+
"grad_norm": 0.40449196100234985,
|
| 4809 |
+
"learning_rate": 5.087840165346194e-06,
|
| 4810 |
+
"loss": 0.8543743133544922,
|
| 4811 |
+
"mean_token_accuracy": 0.8060646489262581,
|
| 4812 |
+
"num_tokens": 10094229.0,
|
| 4813 |
+
"step": 4330
|
| 4814 |
+
},
|
| 4815 |
+
{
|
| 4816 |
+
"entropy": 0.8559481456875802,
|
| 4817 |
+
"epoch": 0.7475347715626749,
|
| 4818 |
+
"grad_norm": 0.46575725078582764,
|
| 4819 |
+
"learning_rate": 5.05339304168102e-06,
|
| 4820 |
+
"loss": 0.8509360313415527,
|
| 4821 |
+
"mean_token_accuracy": 0.8101978525519371,
|
| 4822 |
+
"num_tokens": 10231305.0,
|
| 4823 |
+
"step": 4340
|
| 4824 |
+
},
|
| 4825 |
+
{
|
| 4826 |
+
"entropy": 0.9078745096921921,
|
| 4827 |
+
"epoch": 0.7492572019118977,
|
| 4828 |
+
"grad_norm": 0.4081194996833801,
|
| 4829 |
+
"learning_rate": 5.0189459180158466e-06,
|
| 4830 |
+
"loss": 0.9090764999389649,
|
| 4831 |
+
"mean_token_accuracy": 0.7969685912132263,
|
| 4832 |
+
"num_tokens": 10370440.0,
|
| 4833 |
+
"step": 4350
|
| 4834 |
+
},
|
| 4835 |
+
{
|
| 4836 |
+
"entropy": 0.817448242008686,
|
| 4837 |
+
"epoch": 0.7509796322611204,
|
| 4838 |
+
"grad_norm": 0.3819209337234497,
|
| 4839 |
+
"learning_rate": 4.9844987943506724e-06,
|
| 4840 |
+
"loss": 0.8003293037414551,
|
| 4841 |
+
"mean_token_accuracy": 0.8171394869685173,
|
| 4842 |
+
"num_tokens": 10511468.0,
|
| 4843 |
+
"step": 4360
|
| 4844 |
+
},
|
| 4845 |
+
{
|
| 4846 |
+
"entropy": 0.9328092455863952,
|
| 4847 |
+
"epoch": 0.7527020626103432,
|
| 4848 |
+
"grad_norm": 0.5216546654701233,
|
| 4849 |
+
"learning_rate": 4.950051670685498e-06,
|
| 4850 |
+
"loss": 0.9274360656738281,
|
| 4851 |
+
"mean_token_accuracy": 0.7921751365065575,
|
| 4852 |
+
"num_tokens": 10647454.0,
|
| 4853 |
+
"step": 4370
|
| 4854 |
+
},
|
| 4855 |
+
{
|
| 4856 |
+
"entropy": 0.8536661364138126,
|
| 4857 |
+
"epoch": 0.7544244929595659,
|
| 4858 |
+
"grad_norm": 0.4160046875476837,
|
| 4859 |
+
"learning_rate": 4.915604547020324e-06,
|
| 4860 |
+
"loss": 0.8487312316894531,
|
| 4861 |
+
"mean_token_accuracy": 0.806519165635109,
|
| 4862 |
+
"num_tokens": 10786651.0,
|
| 4863 |
+
"step": 4380
|
| 4864 |
+
},
|
| 4865 |
+
{
|
| 4866 |
+
"entropy": 0.8932063668966294,
|
| 4867 |
+
"epoch": 0.7561469233087887,
|
| 4868 |
+
"grad_norm": 0.5155801177024841,
|
| 4869 |
+
"learning_rate": 4.88115742335515e-06,
|
| 4870 |
+
"loss": 0.8955465316772461,
|
| 4871 |
+
"mean_token_accuracy": 0.7976710572838783,
|
| 4872 |
+
"num_tokens": 10926890.0,
|
| 4873 |
+
"step": 4390
|
| 4874 |
+
},
|
| 4875 |
+
{
|
| 4876 |
+
"entropy": 0.8816228076815605,
|
| 4877 |
+
"epoch": 0.7578693536580114,
|
| 4878 |
+
"grad_norm": 0.4447634220123291,
|
| 4879 |
+
"learning_rate": 4.846710299689976e-06,
|
| 4880 |
+
"loss": 0.8698745727539062,
|
| 4881 |
+
"mean_token_accuracy": 0.8046109586954117,
|
| 4882 |
+
"num_tokens": 11065678.0,
|
| 4883 |
+
"step": 4400
|
| 4884 |
+
},
|
| 4885 |
+
{
|
| 4886 |
+
"epoch": 0.7578693536580114,
|
| 4887 |
+
"eval_entropy": 0.8932004901430765,
|
| 4888 |
+
"eval_loss": 0.8877387642860413,
|
| 4889 |
+
"eval_mean_token_accuracy": 0.8003266229042015,
|
| 4890 |
+
"eval_num_tokens": 11065678.0,
|
| 4891 |
+
"eval_runtime": 1996.1923,
|
| 4892 |
+
"eval_samples_per_second": 5.17,
|
| 4893 |
+
"eval_steps_per_second": 1.293,
|
| 4894 |
+
"step": 4400
|
| 4895 |
+
},
|
| 4896 |
+
{
|
| 4897 |
+
"entropy": 0.8480187527835369,
|
| 4898 |
+
"epoch": 0.7595917840072343,
|
| 4899 |
+
"grad_norm": 0.41860663890838623,
|
| 4900 |
+
"learning_rate": 4.812263176024803e-06,
|
| 4901 |
+
"loss": 0.8360982894897461,
|
| 4902 |
+
"mean_token_accuracy": 0.8108582600951195,
|
| 4903 |
+
"num_tokens": 11202911.0,
|
| 4904 |
+
"step": 4410
|
| 4905 |
+
},
|
| 4906 |
+
{
|
| 4907 |
+
"entropy": 0.880243419110775,
|
| 4908 |
+
"epoch": 0.761314214356457,
|
| 4909 |
+
"grad_norm": 0.4788830280303955,
|
| 4910 |
+
"learning_rate": 4.777816052359629e-06,
|
| 4911 |
+
"loss": 0.8808881759643554,
|
| 4912 |
+
"mean_token_accuracy": 0.7999973684549332,
|
| 4913 |
+
"num_tokens": 11338550.0,
|
| 4914 |
+
"step": 4420
|
| 4915 |
+
},
|
| 4916 |
+
{
|
| 4917 |
+
"entropy": 0.8733625888824463,
|
| 4918 |
+
"epoch": 0.7630366447056797,
|
| 4919 |
+
"grad_norm": 0.5266014337539673,
|
| 4920 |
+
"learning_rate": 4.7433689286944545e-06,
|
| 4921 |
+
"loss": 0.8607860565185547,
|
| 4922 |
+
"mean_token_accuracy": 0.8038028448820114,
|
| 4923 |
+
"num_tokens": 11475542.0,
|
| 4924 |
+
"step": 4430
|
| 4925 |
+
},
|
| 4926 |
+
{
|
| 4927 |
+
"entropy": 0.8758055180311203,
|
| 4928 |
+
"epoch": 0.7647590750549025,
|
| 4929 |
+
"grad_norm": 0.42531901597976685,
|
| 4930 |
+
"learning_rate": 4.70892180502928e-06,
|
| 4931 |
+
"loss": 0.8682390213012695,
|
| 4932 |
+
"mean_token_accuracy": 0.8029620110988617,
|
| 4933 |
+
"num_tokens": 11615072.0,
|
| 4934 |
+
"step": 4440
|
| 4935 |
+
},
|
| 4936 |
+
{
|
| 4937 |
+
"entropy": 0.8809411779046059,
|
| 4938 |
+
"epoch": 0.7664815054041252,
|
| 4939 |
+
"grad_norm": 0.4294584095478058,
|
| 4940 |
+
"learning_rate": 4.674474681364106e-06,
|
| 4941 |
+
"loss": 0.8725706100463867,
|
| 4942 |
+
"mean_token_accuracy": 0.8034122839570046,
|
| 4943 |
+
"num_tokens": 11752076.0,
|
| 4944 |
+
"step": 4450
|
| 4945 |
+
},
|
| 4946 |
+
{
|
| 4947 |
+
"entropy": 0.8637193307280541,
|
| 4948 |
+
"epoch": 0.768203935753348,
|
| 4949 |
+
"grad_norm": 0.4114879071712494,
|
| 4950 |
+
"learning_rate": 4.640027557698932e-06,
|
| 4951 |
+
"loss": 0.8583747863769531,
|
| 4952 |
+
"mean_token_accuracy": 0.8057721823453903,
|
| 4953 |
+
"num_tokens": 11886379.0,
|
| 4954 |
+
"step": 4460
|
| 4955 |
+
},
|
| 4956 |
+
{
|
| 4957 |
+
"entropy": 0.8584412202239037,
|
| 4958 |
+
"epoch": 0.7699263661025707,
|
| 4959 |
+
"grad_norm": 0.42615005373954773,
|
| 4960 |
+
"learning_rate": 4.605580434033758e-06,
|
| 4961 |
+
"loss": 0.8448991775512695,
|
| 4962 |
+
"mean_token_accuracy": 0.8077594563364983,
|
| 4963 |
+
"num_tokens": 12023004.0,
|
| 4964 |
+
"step": 4470
|
| 4965 |
+
},
|
| 4966 |
+
{
|
| 4967 |
+
"entropy": 0.879727041721344,
|
| 4968 |
+
"epoch": 0.7716487964517935,
|
| 4969 |
+
"grad_norm": 0.4471667408943176,
|
| 4970 |
+
"learning_rate": 4.571133310368585e-06,
|
| 4971 |
+
"loss": 0.8692623138427734,
|
| 4972 |
+
"mean_token_accuracy": 0.8025532454252243,
|
| 4973 |
+
"num_tokens": 12159881.0,
|
| 4974 |
+
"step": 4480
|
| 4975 |
+
},
|
| 4976 |
+
{
|
| 4977 |
+
"entropy": 0.8768715851008893,
|
| 4978 |
+
"epoch": 0.7733712268010162,
|
| 4979 |
+
"grad_norm": 0.4854620397090912,
|
| 4980 |
+
"learning_rate": 4.536686186703411e-06,
|
| 4981 |
+
"loss": 0.8687067031860352,
|
| 4982 |
+
"mean_token_accuracy": 0.8022363841533661,
|
| 4983 |
+
"num_tokens": 12304406.0,
|
| 4984 |
+
"step": 4490
|
| 4985 |
+
},
|
| 4986 |
+
{
|
| 4987 |
+
"entropy": 0.882915484905243,
|
| 4988 |
+
"epoch": 0.775093657150239,
|
| 4989 |
+
"grad_norm": 0.48749709129333496,
|
| 4990 |
+
"learning_rate": 4.5022390630382365e-06,
|
| 4991 |
+
"loss": 0.8668482780456543,
|
| 4992 |
+
"mean_token_accuracy": 0.8042670533061027,
|
| 4993 |
+
"num_tokens": 12441894.0,
|
| 4994 |
+
"step": 4500
|
| 4995 |
+
},
|
| 4996 |
+
{
|
| 4997 |
+
"epoch": 0.775093657150239,
|
| 4998 |
+
"eval_entropy": 0.8890064983971125,
|
| 4999 |
+
"eval_loss": 0.887533962726593,
|
| 5000 |
+
"eval_mean_token_accuracy": 0.8003793544706472,
|
| 5001 |
+
"eval_num_tokens": 12441894.0,
|
| 5002 |
+
"eval_runtime": 1998.083,
|
| 5003 |
+
"eval_samples_per_second": 5.165,
|
| 5004 |
+
"eval_steps_per_second": 1.292,
|
| 5005 |
+
"step": 4500
|
| 5006 |
+
},
|
| 5007 |
+
{
|
| 5008 |
+
"entropy": 0.8771491318941116,
|
| 5009 |
+
"epoch": 0.7768160874994617,
|
| 5010 |
+
"grad_norm": 0.45944035053253174,
|
| 5011 |
+
"learning_rate": 4.467791939373063e-06,
|
| 5012 |
+
"loss": 0.8746226310729981,
|
| 5013 |
+
"mean_token_accuracy": 0.8003552362322808,
|
| 5014 |
+
"num_tokens": 12580623.0,
|
| 5015 |
+
"step": 4510
|
| 5016 |
+
},
|
| 5017 |
+
{
|
| 5018 |
+
"entropy": 0.8895734131336213,
|
| 5019 |
+
"epoch": 0.7785385178486844,
|
| 5020 |
+
"grad_norm": 0.4312252104282379,
|
| 5021 |
+
"learning_rate": 4.433344815707889e-06,
|
| 5022 |
+
"loss": 0.8878028869628907,
|
| 5023 |
+
"mean_token_accuracy": 0.800403107702732,
|
| 5024 |
+
"num_tokens": 12716867.0,
|
| 5025 |
+
"step": 4520
|
| 5026 |
+
},
|
| 5027 |
+
{
|
| 5028 |
+
"entropy": 0.8643025323748589,
|
| 5029 |
+
"epoch": 0.7802609481979073,
|
| 5030 |
+
"grad_norm": 0.4234224557876587,
|
| 5031 |
+
"learning_rate": 4.398897692042714e-06,
|
| 5032 |
+
"loss": 0.8600002288818359,
|
| 5033 |
+
"mean_token_accuracy": 0.8056483656167984,
|
| 5034 |
+
"num_tokens": 12853727.0,
|
| 5035 |
+
"step": 4530
|
| 5036 |
+
},
|
| 5037 |
+
{
|
| 5038 |
+
"entropy": 0.8625721678137779,
|
| 5039 |
+
"epoch": 0.78198337854713,
|
| 5040 |
+
"grad_norm": 0.5533012747764587,
|
| 5041 |
+
"learning_rate": 4.364450568377541e-06,
|
| 5042 |
+
"loss": 0.862024974822998,
|
| 5043 |
+
"mean_token_accuracy": 0.8041724354028702,
|
| 5044 |
+
"num_tokens": 12994221.0,
|
| 5045 |
+
"step": 4540
|
| 5046 |
+
},
|
| 5047 |
+
{
|
| 5048 |
+
"entropy": 0.8662130773067475,
|
| 5049 |
+
"epoch": 0.7837058088963528,
|
| 5050 |
+
"grad_norm": 0.39630478620529175,
|
| 5051 |
+
"learning_rate": 4.330003444712367e-06,
|
| 5052 |
+
"loss": 0.860981559753418,
|
| 5053 |
+
"mean_token_accuracy": 0.8042713910341263,
|
| 5054 |
+
"num_tokens": 13130164.0,
|
| 5055 |
+
"step": 4550
|
| 5056 |
+
},
|
| 5057 |
+
{
|
| 5058 |
+
"entropy": 0.911145132780075,
|
| 5059 |
+
"epoch": 0.7854282392455755,
|
| 5060 |
+
"grad_norm": 0.42679402232170105,
|
| 5061 |
+
"learning_rate": 4.295556321047193e-06,
|
| 5062 |
+
"loss": 0.8990155220031738,
|
| 5063 |
+
"mean_token_accuracy": 0.7960023999214172,
|
| 5064 |
+
"num_tokens": 13267850.0,
|
| 5065 |
+
"step": 4560
|
| 5066 |
+
},
|
| 5067 |
+
{
|
| 5068 |
+
"entropy": 0.9082151725888252,
|
| 5069 |
+
"epoch": 0.7871506695947983,
|
| 5070 |
+
"grad_norm": 0.3935401141643524,
|
| 5071 |
+
"learning_rate": 4.261109197382019e-06,
|
| 5072 |
+
"loss": 0.8982461929321289,
|
| 5073 |
+
"mean_token_accuracy": 0.796848364174366,
|
| 5074 |
+
"num_tokens": 13405966.0,
|
| 5075 |
+
"step": 4570
|
| 5076 |
+
},
|
| 5077 |
+
{
|
| 5078 |
+
"entropy": 0.9287943243980408,
|
| 5079 |
+
"epoch": 0.788873099944021,
|
| 5080 |
+
"grad_norm": 0.45502084493637085,
|
| 5081 |
+
"learning_rate": 4.226662073716845e-06,
|
| 5082 |
+
"loss": 0.9211901664733887,
|
| 5083 |
+
"mean_token_accuracy": 0.7938990280032158,
|
| 5084 |
+
"num_tokens": 13541463.0,
|
| 5085 |
+
"step": 4580
|
| 5086 |
+
},
|
| 5087 |
+
{
|
| 5088 |
+
"entropy": 0.8674170449376106,
|
| 5089 |
+
"epoch": 0.7905955302932438,
|
| 5090 |
+
"grad_norm": 0.4692513048648834,
|
| 5091 |
+
"learning_rate": 4.192214950051671e-06,
|
| 5092 |
+
"loss": 0.8526861190795898,
|
| 5093 |
+
"mean_token_accuracy": 0.8061059355735779,
|
| 5094 |
+
"num_tokens": 13679042.0,
|
| 5095 |
+
"step": 4590
|
| 5096 |
+
},
|
| 5097 |
+
{
|
| 5098 |
+
"entropy": 0.8175554417073727,
|
| 5099 |
+
"epoch": 0.7923179606424665,
|
| 5100 |
+
"grad_norm": 0.36742714047431946,
|
| 5101 |
+
"learning_rate": 4.157767826386497e-06,
|
| 5102 |
+
"loss": 0.810401725769043,
|
| 5103 |
+
"mean_token_accuracy": 0.8157213449478149,
|
| 5104 |
+
"num_tokens": 13816805.0,
|
| 5105 |
+
"step": 4600
|
| 5106 |
+
},
|
| 5107 |
+
{
|
| 5108 |
+
"epoch": 0.7923179606424665,
|
| 5109 |
+
"eval_entropy": 0.8860386848588269,
|
| 5110 |
+
"eval_loss": 0.8873475790023804,
|
| 5111 |
+
"eval_mean_token_accuracy": 0.8004140928936115,
|
| 5112 |
+
"eval_num_tokens": 13816805.0,
|
| 5113 |
+
"eval_runtime": 1999.0616,
|
| 5114 |
+
"eval_samples_per_second": 5.163,
|
| 5115 |
+
"eval_steps_per_second": 1.291,
|
| 5116 |
+
"step": 4600
|
| 5117 |
+
},
|
| 5118 |
+
{
|
| 5119 |
+
"entropy": 0.8692925959825516,
|
| 5120 |
+
"epoch": 0.7940403909916893,
|
| 5121 |
+
"grad_norm": 0.43037864565849304,
|
| 5122 |
+
"learning_rate": 4.123320702721323e-06,
|
| 5123 |
+
"loss": 0.87374267578125,
|
| 5124 |
+
"mean_token_accuracy": 0.8035108000040054,
|
| 5125 |
+
"num_tokens": 13956084.0,
|
| 5126 |
+
"step": 4610
|
| 5127 |
+
},
|
| 5128 |
+
{
|
| 5129 |
+
"entropy": 0.9107248976826667,
|
| 5130 |
+
"epoch": 0.795762821340912,
|
| 5131 |
+
"grad_norm": 0.43645474314689636,
|
| 5132 |
+
"learning_rate": 4.088873579056149e-06,
|
| 5133 |
+
"loss": 0.9040485382080078,
|
| 5134 |
+
"mean_token_accuracy": 0.7957431972026825,
|
| 5135 |
+
"num_tokens": 14095417.0,
|
| 5136 |
+
"step": 4620
|
| 5137 |
+
},
|
| 5138 |
+
{
|
| 5139 |
+
"entropy": 0.8810559019446373,
|
| 5140 |
+
"epoch": 0.7974852516901347,
|
| 5141 |
+
"grad_norm": 0.4164002537727356,
|
| 5142 |
+
"learning_rate": 4.054426455390975e-06,
|
| 5143 |
+
"loss": 0.8620301246643066,
|
| 5144 |
+
"mean_token_accuracy": 0.8024208083748817,
|
| 5145 |
+
"num_tokens": 14234163.0,
|
| 5146 |
+
"step": 4630
|
| 5147 |
+
},
|
| 5148 |
+
{
|
| 5149 |
+
"entropy": 0.8721425667405128,
|
| 5150 |
+
"epoch": 0.7992076820393575,
|
| 5151 |
+
"grad_norm": 0.4000791013240814,
|
| 5152 |
+
"learning_rate": 4.0199793317258015e-06,
|
| 5153 |
+
"loss": 0.870361328125,
|
| 5154 |
+
"mean_token_accuracy": 0.8030439510941505,
|
| 5155 |
+
"num_tokens": 14373923.0,
|
| 5156 |
+
"step": 4640
|
| 5157 |
+
},
|
| 5158 |
+
{
|
| 5159 |
+
"entropy": 0.8650285512208938,
|
| 5160 |
+
"epoch": 0.8009301123885803,
|
| 5161 |
+
"grad_norm": 0.49078160524368286,
|
| 5162 |
+
"learning_rate": 3.985532208060627e-06,
|
| 5163 |
+
"loss": 0.8564748764038086,
|
| 5164 |
+
"mean_token_accuracy": 0.8049287557601928,
|
| 5165 |
+
"num_tokens": 14509706.0,
|
| 5166 |
+
"step": 4650
|
| 5167 |
+
},
|
| 5168 |
+
{
|
| 5169 |
+
"entropy": 0.9245395109057426,
|
| 5170 |
+
"epoch": 0.8026525427378031,
|
| 5171 |
+
"grad_norm": 0.4702320694923401,
|
| 5172 |
+
"learning_rate": 3.951085084395453e-06,
|
| 5173 |
+
"loss": 0.9125078201293946,
|
| 5174 |
+
"mean_token_accuracy": 0.7924766167998314,
|
| 5175 |
+
"num_tokens": 14644442.0,
|
| 5176 |
+
"step": 4660
|
| 5177 |
+
},
|
| 5178 |
+
{
|
| 5179 |
+
"entropy": 0.894470140337944,
|
| 5180 |
+
"epoch": 0.8043749730870258,
|
| 5181 |
+
"grad_norm": 0.42101189494132996,
|
| 5182 |
+
"learning_rate": 3.916637960730279e-06,
|
| 5183 |
+
"loss": 0.8770513534545898,
|
| 5184 |
+
"mean_token_accuracy": 0.7988753080368042,
|
| 5185 |
+
"num_tokens": 14782275.0,
|
| 5186 |
+
"step": 4670
|
| 5187 |
+
},
|
| 5188 |
+
{
|
| 5189 |
+
"entropy": 0.829303927719593,
|
| 5190 |
+
"epoch": 0.8060974034362486,
|
| 5191 |
+
"grad_norm": 0.43872424960136414,
|
| 5192 |
+
"learning_rate": 3.882190837065105e-06,
|
| 5193 |
+
"loss": 0.8163064956665039,
|
| 5194 |
+
"mean_token_accuracy": 0.8142339214682579,
|
| 5195 |
+
"num_tokens": 14922228.0,
|
| 5196 |
+
"step": 4680
|
| 5197 |
+
},
|
| 5198 |
+
{
|
| 5199 |
+
"entropy": 0.8826430246233941,
|
| 5200 |
+
"epoch": 0.8078198337854713,
|
| 5201 |
+
"grad_norm": 0.5022356510162354,
|
| 5202 |
+
"learning_rate": 3.847743713399931e-06,
|
| 5203 |
+
"loss": 0.8792767524719238,
|
| 5204 |
+
"mean_token_accuracy": 0.8019203156232834,
|
| 5205 |
+
"num_tokens": 15058860.0,
|
| 5206 |
+
"step": 4690
|
| 5207 |
+
},
|
| 5208 |
+
{
|
| 5209 |
+
"entropy": 0.8542191177606583,
|
| 5210 |
+
"epoch": 0.8095422641346941,
|
| 5211 |
+
"grad_norm": 0.39593997597694397,
|
| 5212 |
+
"learning_rate": 3.8132965897347572e-06,
|
| 5213 |
+
"loss": 0.8465509414672852,
|
| 5214 |
+
"mean_token_accuracy": 0.8096876740455627,
|
| 5215 |
+
"num_tokens": 15197803.0,
|
| 5216 |
+
"step": 4700
|
| 5217 |
+
},
|
| 5218 |
+
{
|
| 5219 |
+
"epoch": 0.8095422641346941,
|
| 5220 |
+
"eval_entropy": 0.8892219523411765,
|
| 5221 |
+
"eval_loss": 0.8871769905090332,
|
| 5222 |
+
"eval_mean_token_accuracy": 0.8004685473368178,
|
| 5223 |
+
"eval_num_tokens": 15197803.0,
|
| 5224 |
+
"eval_runtime": 1999.6905,
|
| 5225 |
+
"eval_samples_per_second": 5.161,
|
| 5226 |
+
"eval_steps_per_second": 1.291,
|
| 5227 |
+
"step": 4700
|
| 5228 |
+
},
|
| 5229 |
+
{
|
| 5230 |
+
"entropy": 0.8803235799074173,
|
| 5231 |
+
"epoch": 0.8112646944839168,
|
| 5232 |
+
"grad_norm": 0.4291439652442932,
|
| 5233 |
+
"learning_rate": 3.7788494660695836e-06,
|
| 5234 |
+
"loss": 0.8776756286621094,
|
| 5235 |
+
"mean_token_accuracy": 0.8018289491534233,
|
| 5236 |
+
"num_tokens": 15333287.0,
|
| 5237 |
+
"step": 4710
|
| 5238 |
+
},
|
| 5239 |
+
{
|
| 5240 |
+
"entropy": 0.8909836888313294,
|
| 5241 |
+
"epoch": 0.8129871248331395,
|
| 5242 |
+
"grad_norm": 0.429242342710495,
|
| 5243 |
+
"learning_rate": 3.7444023424044094e-06,
|
| 5244 |
+
"loss": 0.8833690643310547,
|
| 5245 |
+
"mean_token_accuracy": 0.7992347851395607,
|
| 5246 |
+
"num_tokens": 15469520.0,
|
| 5247 |
+
"step": 4720
|
| 5248 |
+
},
|
| 5249 |
+
{
|
| 5250 |
+
"entropy": 0.8946880295872688,
|
| 5251 |
+
"epoch": 0.8147095551823623,
|
| 5252 |
+
"grad_norm": 0.5099314451217651,
|
| 5253 |
+
"learning_rate": 3.7099552187392357e-06,
|
| 5254 |
+
"loss": 0.8901423454284668,
|
| 5255 |
+
"mean_token_accuracy": 0.7986397400498391,
|
| 5256 |
+
"num_tokens": 15607616.0,
|
| 5257 |
+
"step": 4730
|
| 5258 |
+
},
|
| 5259 |
+
{
|
| 5260 |
+
"entropy": 0.9065332308411598,
|
| 5261 |
+
"epoch": 0.816431985531585,
|
| 5262 |
+
"grad_norm": 0.4736480116844177,
|
| 5263 |
+
"learning_rate": 3.6755080950740616e-06,
|
| 5264 |
+
"loss": 0.9019441604614258,
|
| 5265 |
+
"mean_token_accuracy": 0.797715176641941,
|
| 5266 |
+
"num_tokens": 15744067.0,
|
| 5267 |
+
"step": 4740
|
| 5268 |
+
},
|
| 5269 |
+
{
|
| 5270 |
+
"entropy": 0.8654450118541718,
|
| 5271 |
+
"epoch": 0.8181544158808078,
|
| 5272 |
+
"grad_norm": 0.4694020748138428,
|
| 5273 |
+
"learning_rate": 3.641060971408888e-06,
|
| 5274 |
+
"loss": 0.8600435256958008,
|
| 5275 |
+
"mean_token_accuracy": 0.8058715596795082,
|
| 5276 |
+
"num_tokens": 15883159.0,
|
| 5277 |
+
"step": 4750
|
| 5278 |
+
},
|
| 5279 |
+
{
|
| 5280 |
+
"entropy": 0.8759298019111157,
|
| 5281 |
+
"epoch": 0.8198768462300305,
|
| 5282 |
+
"grad_norm": 0.5040362477302551,
|
| 5283 |
+
"learning_rate": 3.6066138477437134e-06,
|
| 5284 |
+
"loss": 0.8586697578430176,
|
| 5285 |
+
"mean_token_accuracy": 0.806798329949379,
|
| 5286 |
+
"num_tokens": 16021725.0,
|
| 5287 |
+
"step": 4760
|
| 5288 |
+
},
|
| 5289 |
+
{
|
| 5290 |
+
"entropy": 0.8462592497467994,
|
| 5291 |
+
"epoch": 0.8215992765792534,
|
| 5292 |
+
"grad_norm": 0.4179896414279938,
|
| 5293 |
+
"learning_rate": 3.5721667240785397e-06,
|
| 5294 |
+
"loss": 0.8351809501647949,
|
| 5295 |
+
"mean_token_accuracy": 0.8095413893461227,
|
| 5296 |
+
"num_tokens": 16157037.0,
|
| 5297 |
+
"step": 4770
|
| 5298 |
+
},
|
| 5299 |
+
{
|
| 5300 |
+
"entropy": 0.8456560544669628,
|
| 5301 |
+
"epoch": 0.8233217069284761,
|
| 5302 |
+
"grad_norm": 0.4607621431350708,
|
| 5303 |
+
"learning_rate": 3.5377196004133656e-06,
|
| 5304 |
+
"loss": 0.8377568244934082,
|
| 5305 |
+
"mean_token_accuracy": 0.8099890291690827,
|
| 5306 |
+
"num_tokens": 16294647.0,
|
| 5307 |
+
"step": 4780
|
| 5308 |
+
},
|
| 5309 |
+
{
|
| 5310 |
+
"entropy": 0.8446394450962543,
|
| 5311 |
+
"epoch": 0.8250441372776989,
|
| 5312 |
+
"grad_norm": 0.41511309146881104,
|
| 5313 |
+
"learning_rate": 3.503272476748192e-06,
|
| 5314 |
+
"loss": 0.8328559875488282,
|
| 5315 |
+
"mean_token_accuracy": 0.8099060103297233,
|
| 5316 |
+
"num_tokens": 16431574.0,
|
| 5317 |
+
"step": 4790
|
| 5318 |
+
},
|
| 5319 |
+
{
|
| 5320 |
+
"entropy": 0.8220671258866787,
|
| 5321 |
+
"epoch": 0.8267665676269216,
|
| 5322 |
+
"grad_norm": 0.41917988657951355,
|
| 5323 |
+
"learning_rate": 3.468825353083018e-06,
|
| 5324 |
+
"loss": 0.8178951263427734,
|
| 5325 |
+
"mean_token_accuracy": 0.815338246524334,
|
| 5326 |
+
"num_tokens": 16573221.0,
|
| 5327 |
+
"step": 4800
|
| 5328 |
+
},
|
| 5329 |
+
{
|
| 5330 |
+
"epoch": 0.8267665676269216,
|
| 5331 |
+
"eval_entropy": 0.889032811815542,
|
| 5332 |
+
"eval_loss": 0.8869978785514832,
|
| 5333 |
+
"eval_mean_token_accuracy": 0.8004639150559648,
|
| 5334 |
+
"eval_num_tokens": 16573221.0,
|
| 5335 |
+
"eval_runtime": 1999.6311,
|
| 5336 |
+
"eval_samples_per_second": 5.161,
|
| 5337 |
+
"eval_steps_per_second": 1.291,
|
| 5338 |
+
"step": 4800
|
| 5339 |
}
|
| 5340 |
],
|
| 5341 |
"logging_steps": 10,
|
|
|
|
| 5355 |
"attributes": {}
|
| 5356 |
}
|
| 5357 |
},
|
| 5358 |
+
"total_flos": 7.887012992552796e+17,
|
| 5359 |
"train_batch_size": 2,
|
| 5360 |
"trial_name": null,
|
| 5361 |
"trial_params": null
|