Training in progress, step 8500, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -30,15 +30,15 @@
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
| 33 |
-
"
|
| 34 |
"in_proj",
|
|
|
|
|
|
|
|
|
|
| 35 |
"v_proj",
|
| 36 |
-
"out_proj",
|
| 37 |
-
"w2",
|
| 38 |
-
"w3",
|
| 39 |
"o_proj",
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
],
|
| 43 |
"target_parameters": null,
|
| 44 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
| 33 |
+
"w2",
|
| 34 |
"in_proj",
|
| 35 |
+
"q_proj",
|
| 36 |
+
"w1",
|
| 37 |
+
"k_proj",
|
| 38 |
"v_proj",
|
|
|
|
|
|
|
|
|
|
| 39 |
"o_proj",
|
| 40 |
+
"out_proj",
|
| 41 |
+
"w3"
|
| 42 |
],
|
| 43 |
"target_parameters": null,
|
| 44 |
"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 88891320
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d52fa8d7add313da353fe1f64f6707d0c28817e67e9aed531b5a4fb07525e57
|
| 3 |
size 88891320
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 177891851
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4f2eabea51c210fd332a72f892bfa3243ef82131d28a92e3c4589de8c592c7d9
|
| 3 |
size 177891851
|
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:df6ee5cdefa4ee4746cb89edc007236a7fdf4cfa3192ff448175f602c35556c8
|
| 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:a6475dc6dbd12c7bdd9ac559042d6443bbc2b477cb550c9bc0bf4e7a109e2d09
|
| 3 |
size 14917
|
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 sha256:911e304d28fbd8d316a961f622b76acc9eb9626d1cebb6fdd73bec5b7ccaf535
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "./lfm25-kodcode-lora/checkpoint-
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 250,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -7838,6 +7838,1050 @@
|
|
| 7838 |
"eval_samples_per_second": 5.968,
|
| 7839 |
"eval_steps_per_second": 1.493,
|
| 7840 |
"step": 7500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7841 |
}
|
| 7842 |
],
|
| 7843 |
"logging_steps": 10,
|
|
@@ -7857,7 +8901,7 @@
|
|
| 7857 |
"attributes": {}
|
| 7858 |
}
|
| 7859 |
},
|
| 7860 |
-
"total_flos":
|
| 7861 |
"train_batch_size": 2,
|
| 7862 |
"trial_name": null,
|
| 7863 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 8500,
|
| 3 |
+
"best_metric": 0.29068824648857117,
|
| 4 |
+
"best_model_checkpoint": "./lfm25-kodcode-lora/checkpoint-8500",
|
| 5 |
+
"epoch": 0.6439515899922347,
|
| 6 |
"eval_steps": 250,
|
| 7 |
+
"global_step": 8500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 7838 |
"eval_samples_per_second": 5.968,
|
| 7839 |
"eval_steps_per_second": 1.493,
|
| 7840 |
"step": 7500
|
| 7841 |
+
},
|
| 7842 |
+
{
|
| 7843 |
+
"entropy": 0.3109708122909069,
|
| 7844 |
+
"epoch": 0.5689501695107861,
|
| 7845 |
+
"grad_norm": 0.4822841286659241,
|
| 7846 |
+
"learning_rate": 4.132161835697846e-05,
|
| 7847 |
+
"loss": 0.3667874336242676,
|
| 7848 |
+
"mean_token_accuracy": 0.9005939975380898,
|
| 7849 |
+
"num_tokens": 114968.0,
|
| 7850 |
+
"step": 7510
|
| 7851 |
+
},
|
| 7852 |
+
{
|
| 7853 |
+
"entropy": 0.3276409078389406,
|
| 7854 |
+
"epoch": 0.5697077596166594,
|
| 7855 |
+
"grad_norm": 0.4089864492416382,
|
| 7856 |
+
"learning_rate": 4.120082652019689e-05,
|
| 7857 |
+
"loss": 0.38182897567749025,
|
| 7858 |
+
"mean_token_accuracy": 0.8978955864906311,
|
| 7859 |
+
"num_tokens": 224970.0,
|
| 7860 |
+
"step": 7520
|
| 7861 |
+
},
|
| 7862 |
+
{
|
| 7863 |
+
"entropy": 0.2703943993896246,
|
| 7864 |
+
"epoch": 0.5704653497225326,
|
| 7865 |
+
"grad_norm": 0.39988771080970764,
|
| 7866 |
+
"learning_rate": 4.108008765586738e-05,
|
| 7867 |
+
"loss": 0.3071948528289795,
|
| 7868 |
+
"mean_token_accuracy": 0.9152684897184372,
|
| 7869 |
+
"num_tokens": 324147.0,
|
| 7870 |
+
"step": 7530
|
| 7871 |
+
},
|
| 7872 |
+
{
|
| 7873 |
+
"entropy": 0.2900085491128266,
|
| 7874 |
+
"epoch": 0.5712229398284059,
|
| 7875 |
+
"grad_norm": 0.40979936718940735,
|
| 7876 |
+
"learning_rate": 4.095940249085749e-05,
|
| 7877 |
+
"loss": 0.3266597270965576,
|
| 7878 |
+
"mean_token_accuracy": 0.9085231080651284,
|
| 7879 |
+
"num_tokens": 421752.0,
|
| 7880 |
+
"step": 7540
|
| 7881 |
+
},
|
| 7882 |
+
{
|
| 7883 |
+
"entropy": 0.2737187416292727,
|
| 7884 |
+
"epoch": 0.5719805299342791,
|
| 7885 |
+
"grad_norm": 0.4078211784362793,
|
| 7886 |
+
"learning_rate": 4.0838771751711505e-05,
|
| 7887 |
+
"loss": 0.3361893892288208,
|
| 7888 |
+
"mean_token_accuracy": 0.9113953113555908,
|
| 7889 |
+
"num_tokens": 517042.0,
|
| 7890 |
+
"step": 7550
|
| 7891 |
+
},
|
| 7892 |
+
{
|
| 7893 |
+
"entropy": 0.31109195090830327,
|
| 7894 |
+
"epoch": 0.5727381200401522,
|
| 7895 |
+
"grad_norm": 0.406087189912796,
|
| 7896 |
+
"learning_rate": 4.0718196164646036e-05,
|
| 7897 |
+
"loss": 0.389078950881958,
|
| 7898 |
+
"mean_token_accuracy": 0.8990072026848793,
|
| 7899 |
+
"num_tokens": 621106.0,
|
| 7900 |
+
"step": 7560
|
| 7901 |
+
},
|
| 7902 |
+
{
|
| 7903 |
+
"entropy": 0.2715472083538771,
|
| 7904 |
+
"epoch": 0.5734957101460255,
|
| 7905 |
+
"grad_norm": 0.48067522048950195,
|
| 7906 |
+
"learning_rate": 4.059767645554568e-05,
|
| 7907 |
+
"loss": 0.3081014633178711,
|
| 7908 |
+
"mean_token_accuracy": 0.914455634355545,
|
| 7909 |
+
"num_tokens": 720233.0,
|
| 7910 |
+
"step": 7570
|
| 7911 |
+
},
|
| 7912 |
+
{
|
| 7913 |
+
"entropy": 0.32704932652413843,
|
| 7914 |
+
"epoch": 0.5742533002518987,
|
| 7915 |
+
"grad_norm": 0.5467900633811951,
|
| 7916 |
+
"learning_rate": 4.047721334995865e-05,
|
| 7917 |
+
"loss": 0.376554012298584,
|
| 7918 |
+
"mean_token_accuracy": 0.8963387101888657,
|
| 7919 |
+
"num_tokens": 818271.0,
|
| 7920 |
+
"step": 7580
|
| 7921 |
+
},
|
| 7922 |
+
{
|
| 7923 |
+
"entropy": 0.3027558699250221,
|
| 7924 |
+
"epoch": 0.575010890357772,
|
| 7925 |
+
"grad_norm": 0.5470854640007019,
|
| 7926 |
+
"learning_rate": 4.035680757309238e-05,
|
| 7927 |
+
"loss": 0.34917449951171875,
|
| 7928 |
+
"mean_token_accuracy": 0.9045404583215714,
|
| 7929 |
+
"num_tokens": 918405.0,
|
| 7930 |
+
"step": 7590
|
| 7931 |
+
},
|
| 7932 |
+
{
|
| 7933 |
+
"entropy": 0.29923666920512915,
|
| 7934 |
+
"epoch": 0.5757684804636451,
|
| 7935 |
+
"grad_norm": 0.5230236053466797,
|
| 7936 |
+
"learning_rate": 4.023645984980919e-05,
|
| 7937 |
+
"loss": 0.34145679473876955,
|
| 7938 |
+
"mean_token_accuracy": 0.9045537799596787,
|
| 7939 |
+
"num_tokens": 1018239.0,
|
| 7940 |
+
"step": 7600
|
| 7941 |
+
},
|
| 7942 |
+
{
|
| 7943 |
+
"entropy": 0.31514527965337036,
|
| 7944 |
+
"epoch": 0.5765260705695183,
|
| 7945 |
+
"grad_norm": 0.3059269189834595,
|
| 7946 |
+
"learning_rate": 4.0116170904621896e-05,
|
| 7947 |
+
"loss": 0.37950544357299804,
|
| 7948 |
+
"mean_token_accuracy": 0.8975080713629723,
|
| 7949 |
+
"num_tokens": 1116705.0,
|
| 7950 |
+
"step": 7610
|
| 7951 |
+
},
|
| 7952 |
+
{
|
| 7953 |
+
"entropy": 0.26913794968277216,
|
| 7954 |
+
"epoch": 0.5772836606753916,
|
| 7955 |
+
"grad_norm": 0.5136588215827942,
|
| 7956 |
+
"learning_rate": 3.999594146168947e-05,
|
| 7957 |
+
"loss": 0.3233908176422119,
|
| 7958 |
+
"mean_token_accuracy": 0.9152696698904037,
|
| 7959 |
+
"num_tokens": 1213766.0,
|
| 7960 |
+
"step": 7620
|
| 7961 |
+
},
|
| 7962 |
+
{
|
| 7963 |
+
"entropy": 0.33589745350182054,
|
| 7964 |
+
"epoch": 0.5780412507812648,
|
| 7965 |
+
"grad_norm": 0.4772613048553467,
|
| 7966 |
+
"learning_rate": 3.987577224481266e-05,
|
| 7967 |
+
"loss": 0.3794508934020996,
|
| 7968 |
+
"mean_token_accuracy": 0.8917440369725227,
|
| 7969 |
+
"num_tokens": 1321188.0,
|
| 7970 |
+
"step": 7630
|
| 7971 |
+
},
|
| 7972 |
+
{
|
| 7973 |
+
"entropy": 0.2889779403805733,
|
| 7974 |
+
"epoch": 0.5787988408871381,
|
| 7975 |
+
"grad_norm": 0.36705854535102844,
|
| 7976 |
+
"learning_rate": 3.975566397742965e-05,
|
| 7977 |
+
"loss": 0.3197050094604492,
|
| 7978 |
+
"mean_token_accuracy": 0.9126046910881996,
|
| 7979 |
+
"num_tokens": 1417615.0,
|
| 7980 |
+
"step": 7640
|
| 7981 |
+
},
|
| 7982 |
+
{
|
| 7983 |
+
"entropy": 0.3254348419606686,
|
| 7984 |
+
"epoch": 0.5795564309930112,
|
| 7985 |
+
"grad_norm": 0.3900688886642456,
|
| 7986 |
+
"learning_rate": 3.963561738261173e-05,
|
| 7987 |
+
"loss": 0.37628955841064454,
|
| 7988 |
+
"mean_token_accuracy": 0.8971853539347648,
|
| 7989 |
+
"num_tokens": 1515666.0,
|
| 7990 |
+
"step": 7650
|
| 7991 |
+
},
|
| 7992 |
+
{
|
| 7993 |
+
"entropy": 0.29808914177119733,
|
| 7994 |
+
"epoch": 0.5803140210988844,
|
| 7995 |
+
"grad_norm": 0.4222153127193451,
|
| 7996 |
+
"learning_rate": 3.951563318305883e-05,
|
| 7997 |
+
"loss": 0.33919458389282225,
|
| 7998 |
+
"mean_token_accuracy": 0.9053486153483391,
|
| 7999 |
+
"num_tokens": 1616529.0,
|
| 8000 |
+
"step": 7660
|
| 8001 |
+
},
|
| 8002 |
+
{
|
| 8003 |
+
"entropy": 0.30099980514496566,
|
| 8004 |
+
"epoch": 0.5810716112047577,
|
| 8005 |
+
"grad_norm": 0.4373079538345337,
|
| 8006 |
+
"learning_rate": 3.939571210109532e-05,
|
| 8007 |
+
"loss": 0.3660151958465576,
|
| 8008 |
+
"mean_token_accuracy": 0.9015607729554176,
|
| 8009 |
+
"num_tokens": 1711473.0,
|
| 8010 |
+
"step": 7670
|
| 8011 |
+
},
|
| 8012 |
+
{
|
| 8013 |
+
"entropy": 0.25706873182207346,
|
| 8014 |
+
"epoch": 0.5818292013106309,
|
| 8015 |
+
"grad_norm": 0.49894335865974426,
|
| 8016 |
+
"learning_rate": 3.9275854858665564e-05,
|
| 8017 |
+
"loss": 0.3213352680206299,
|
| 8018 |
+
"mean_token_accuracy": 0.9181353032588959,
|
| 8019 |
+
"num_tokens": 1794864.0,
|
| 8020 |
+
"step": 7680
|
| 8021 |
+
},
|
| 8022 |
+
{
|
| 8023 |
+
"entropy": 0.29280857387930154,
|
| 8024 |
+
"epoch": 0.582586791416504,
|
| 8025 |
+
"grad_norm": 0.35843634605407715,
|
| 8026 |
+
"learning_rate": 3.915606217732962e-05,
|
| 8027 |
+
"loss": 0.32027444839477537,
|
| 8028 |
+
"mean_token_accuracy": 0.9076616898179054,
|
| 8029 |
+
"num_tokens": 1889475.0,
|
| 8030 |
+
"step": 7690
|
| 8031 |
+
},
|
| 8032 |
+
{
|
| 8033 |
+
"entropy": 0.2827717169187963,
|
| 8034 |
+
"epoch": 0.5833443815223773,
|
| 8035 |
+
"grad_norm": 0.41261976957321167,
|
| 8036 |
+
"learning_rate": 3.903633477825886e-05,
|
| 8037 |
+
"loss": 0.33794167041778567,
|
| 8038 |
+
"mean_token_accuracy": 0.9098512396216393,
|
| 8039 |
+
"num_tokens": 1992584.0,
|
| 8040 |
+
"step": 7700
|
| 8041 |
+
},
|
| 8042 |
+
{
|
| 8043 |
+
"entropy": 0.33711165077984334,
|
| 8044 |
+
"epoch": 0.5841019716282505,
|
| 8045 |
+
"grad_norm": 0.44373470544815063,
|
| 8046 |
+
"learning_rate": 3.891667338223167e-05,
|
| 8047 |
+
"loss": 0.381363582611084,
|
| 8048 |
+
"mean_token_accuracy": 0.8914385855197906,
|
| 8049 |
+
"num_tokens": 2109147.0,
|
| 8050 |
+
"step": 7710
|
| 8051 |
+
},
|
| 8052 |
+
{
|
| 8053 |
+
"entropy": 0.2898610128089786,
|
| 8054 |
+
"epoch": 0.5848595617341238,
|
| 8055 |
+
"grad_norm": 0.47294947504997253,
|
| 8056 |
+
"learning_rate": 3.879707870962904e-05,
|
| 8057 |
+
"loss": 0.37704427242279054,
|
| 8058 |
+
"mean_token_accuracy": 0.9094848409295082,
|
| 8059 |
+
"num_tokens": 2206523.0,
|
| 8060 |
+
"step": 7720
|
| 8061 |
+
},
|
| 8062 |
+
{
|
| 8063 |
+
"entropy": 0.3302390962839127,
|
| 8064 |
+
"epoch": 0.585617151839997,
|
| 8065 |
+
"grad_norm": 0.4247268736362457,
|
| 8066 |
+
"learning_rate": 3.867755148043032e-05,
|
| 8067 |
+
"loss": 0.36937775611877444,
|
| 8068 |
+
"mean_token_accuracy": 0.8964511677622795,
|
| 8069 |
+
"num_tokens": 2313042.0,
|
| 8070 |
+
"step": 7730
|
| 8071 |
+
},
|
| 8072 |
+
{
|
| 8073 |
+
"entropy": 0.29674111772328615,
|
| 8074 |
+
"epoch": 0.5863747419458701,
|
| 8075 |
+
"grad_norm": 0.38928377628326416,
|
| 8076 |
+
"learning_rate": 3.855809241420883e-05,
|
| 8077 |
+
"loss": 0.35187516212463377,
|
| 8078 |
+
"mean_token_accuracy": 0.907732842862606,
|
| 8079 |
+
"num_tokens": 2412624.0,
|
| 8080 |
+
"step": 7740
|
| 8081 |
+
},
|
| 8082 |
+
{
|
| 8083 |
+
"entropy": 0.26236414816230536,
|
| 8084 |
+
"epoch": 0.5871323320517434,
|
| 8085 |
+
"grad_norm": 0.44227948784828186,
|
| 8086 |
+
"learning_rate": 3.843870223012753e-05,
|
| 8087 |
+
"loss": 0.3066854000091553,
|
| 8088 |
+
"mean_token_accuracy": 0.918500104546547,
|
| 8089 |
+
"num_tokens": 2502901.0,
|
| 8090 |
+
"step": 7750
|
| 8091 |
+
},
|
| 8092 |
+
{
|
| 8093 |
+
"epoch": 0.5871323320517434,
|
| 8094 |
+
"eval_entropy": 0.29141705292050313,
|
| 8095 |
+
"eval_loss": 0.29264143109321594,
|
| 8096 |
+
"eval_mean_token_accuracy": 0.9052422897199567,
|
| 8097 |
+
"eval_num_tokens": 2502901.0,
|
| 8098 |
+
"eval_runtime": 366.9036,
|
| 8099 |
+
"eval_samples_per_second": 5.816,
|
| 8100 |
+
"eval_steps_per_second": 1.455,
|
| 8101 |
+
"step": 7750
|
| 8102 |
+
},
|
| 8103 |
+
{
|
| 8104 |
+
"entropy": 0.29118088809773324,
|
| 8105 |
+
"epoch": 0.5878899221576166,
|
| 8106 |
+
"grad_norm": 0.40471747517585754,
|
| 8107 |
+
"learning_rate": 3.831938164693469e-05,
|
| 8108 |
+
"loss": 0.3376481056213379,
|
| 8109 |
+
"mean_token_accuracy": 0.9038581416010857,
|
| 8110 |
+
"num_tokens": 2603072.0,
|
| 8111 |
+
"step": 7760
|
| 8112 |
+
},
|
| 8113 |
+
{
|
| 8114 |
+
"entropy": 0.3052817082032561,
|
| 8115 |
+
"epoch": 0.5886475122634899,
|
| 8116 |
+
"grad_norm": 0.5422359704971313,
|
| 8117 |
+
"learning_rate": 3.8200131382959586e-05,
|
| 8118 |
+
"loss": 0.34176883697509763,
|
| 8119 |
+
"mean_token_accuracy": 0.9007858261466026,
|
| 8120 |
+
"num_tokens": 2705340.0,
|
| 8121 |
+
"step": 7770
|
| 8122 |
+
},
|
| 8123 |
+
{
|
| 8124 |
+
"entropy": 0.2885507680475712,
|
| 8125 |
+
"epoch": 0.589405102369363,
|
| 8126 |
+
"grad_norm": 0.5075405240058899,
|
| 8127 |
+
"learning_rate": 3.808095215610816e-05,
|
| 8128 |
+
"loss": 0.3256500720977783,
|
| 8129 |
+
"mean_token_accuracy": 0.9076829209923745,
|
| 8130 |
+
"num_tokens": 2801725.0,
|
| 8131 |
+
"step": 7780
|
| 8132 |
+
},
|
| 8133 |
+
{
|
| 8134 |
+
"entropy": 0.3176051995716989,
|
| 8135 |
+
"epoch": 0.5901626924752362,
|
| 8136 |
+
"grad_norm": 0.36726605892181396,
|
| 8137 |
+
"learning_rate": 3.7961844683858684e-05,
|
| 8138 |
+
"loss": 0.36698002815246583,
|
| 8139 |
+
"mean_token_accuracy": 0.8985909134149551,
|
| 8140 |
+
"num_tokens": 2901989.0,
|
| 8141 |
+
"step": 7790
|
| 8142 |
+
},
|
| 8143 |
+
{
|
| 8144 |
+
"entropy": 0.3000574687495828,
|
| 8145 |
+
"epoch": 0.5909202825811095,
|
| 8146 |
+
"grad_norm": 0.5310767889022827,
|
| 8147 |
+
"learning_rate": 3.784280968325748e-05,
|
| 8148 |
+
"loss": 0.3372817039489746,
|
| 8149 |
+
"mean_token_accuracy": 0.9080954909324646,
|
| 8150 |
+
"num_tokens": 2998881.0,
|
| 8151 |
+
"step": 7800
|
| 8152 |
+
},
|
| 8153 |
+
{
|
| 8154 |
+
"entropy": 0.2850060695782304,
|
| 8155 |
+
"epoch": 0.5916778726869827,
|
| 8156 |
+
"grad_norm": 0.39137741923332214,
|
| 8157 |
+
"learning_rate": 3.7723847870914575e-05,
|
| 8158 |
+
"loss": 0.34608912467956543,
|
| 8159 |
+
"mean_token_accuracy": 0.9057006880640983,
|
| 8160 |
+
"num_tokens": 3099067.0,
|
| 8161 |
+
"step": 7810
|
| 8162 |
+
},
|
| 8163 |
+
{
|
| 8164 |
+
"entropy": 0.28733792770653965,
|
| 8165 |
+
"epoch": 0.592435462792856,
|
| 8166 |
+
"grad_norm": 0.3380966782569885,
|
| 8167 |
+
"learning_rate": 3.760495996299935e-05,
|
| 8168 |
+
"loss": 0.32074360847473143,
|
| 8169 |
+
"mean_token_accuracy": 0.907772421836853,
|
| 8170 |
+
"num_tokens": 3199784.0,
|
| 8171 |
+
"step": 7820
|
| 8172 |
+
},
|
| 8173 |
+
{
|
| 8174 |
+
"entropy": 0.3030220426619053,
|
| 8175 |
+
"epoch": 0.5931930528987291,
|
| 8176 |
+
"grad_norm": 0.5010019540786743,
|
| 8177 |
+
"learning_rate": 3.748614667523632e-05,
|
| 8178 |
+
"loss": 0.3495840072631836,
|
| 8179 |
+
"mean_token_accuracy": 0.9055353298783302,
|
| 8180 |
+
"num_tokens": 3301790.0,
|
| 8181 |
+
"step": 7830
|
| 8182 |
+
},
|
| 8183 |
+
{
|
| 8184 |
+
"entropy": 0.25674657318741084,
|
| 8185 |
+
"epoch": 0.5939506430046023,
|
| 8186 |
+
"grad_norm": 0.5346329212188721,
|
| 8187 |
+
"learning_rate": 3.736740872290076e-05,
|
| 8188 |
+
"loss": 0.3107120513916016,
|
| 8189 |
+
"mean_token_accuracy": 0.9172234773635864,
|
| 8190 |
+
"num_tokens": 3391263.0,
|
| 8191 |
+
"step": 7840
|
| 8192 |
+
},
|
| 8193 |
+
{
|
| 8194 |
+
"entropy": 0.3097769618034363,
|
| 8195 |
+
"epoch": 0.5947082331104756,
|
| 8196 |
+
"grad_norm": 0.5312048196792603,
|
| 8197 |
+
"learning_rate": 3.7248746820814405e-05,
|
| 8198 |
+
"loss": 0.3648669719696045,
|
| 8199 |
+
"mean_token_accuracy": 0.898509131371975,
|
| 8200 |
+
"num_tokens": 3493351.0,
|
| 8201 |
+
"step": 7850
|
| 8202 |
+
},
|
| 8203 |
+
{
|
| 8204 |
+
"entropy": 0.3084961650893092,
|
| 8205 |
+
"epoch": 0.5954658232163488,
|
| 8206 |
+
"grad_norm": 0.38531121611595154,
|
| 8207 |
+
"learning_rate": 3.7130161683341146e-05,
|
| 8208 |
+
"loss": 0.36528897285461426,
|
| 8209 |
+
"mean_token_accuracy": 0.9024834498763085,
|
| 8210 |
+
"num_tokens": 3596548.0,
|
| 8211 |
+
"step": 7860
|
| 8212 |
+
},
|
| 8213 |
+
{
|
| 8214 |
+
"entropy": 0.28346777241677046,
|
| 8215 |
+
"epoch": 0.5962234133222221,
|
| 8216 |
+
"grad_norm": 0.39585480093955994,
|
| 8217 |
+
"learning_rate": 3.701165402438277e-05,
|
| 8218 |
+
"loss": 0.3357419967651367,
|
| 8219 |
+
"mean_token_accuracy": 0.9120162844657898,
|
| 8220 |
+
"num_tokens": 3695360.0,
|
| 8221 |
+
"step": 7870
|
| 8222 |
+
},
|
| 8223 |
+
{
|
| 8224 |
+
"entropy": 0.3283775428310037,
|
| 8225 |
+
"epoch": 0.5969810034280952,
|
| 8226 |
+
"grad_norm": 0.2965605854988098,
|
| 8227 |
+
"learning_rate": 3.6893224557374595e-05,
|
| 8228 |
+
"loss": 0.37510778903961184,
|
| 8229 |
+
"mean_token_accuracy": 0.8940520346164703,
|
| 8230 |
+
"num_tokens": 3795712.0,
|
| 8231 |
+
"step": 7880
|
| 8232 |
+
},
|
| 8233 |
+
{
|
| 8234 |
+
"entropy": 0.303850819543004,
|
| 8235 |
+
"epoch": 0.5977385935339684,
|
| 8236 |
+
"grad_norm": 0.5088040828704834,
|
| 8237 |
+
"learning_rate": 3.677487399528124e-05,
|
| 8238 |
+
"loss": 0.32466897964477537,
|
| 8239 |
+
"mean_token_accuracy": 0.9058476537466049,
|
| 8240 |
+
"num_tokens": 3891781.0,
|
| 8241 |
+
"step": 7890
|
| 8242 |
+
},
|
| 8243 |
+
{
|
| 8244 |
+
"entropy": 0.28571315221488475,
|
| 8245 |
+
"epoch": 0.5984961836398417,
|
| 8246 |
+
"grad_norm": 0.39025136828422546,
|
| 8247 |
+
"learning_rate": 3.6656603050592294e-05,
|
| 8248 |
+
"loss": 0.3613213062286377,
|
| 8249 |
+
"mean_token_accuracy": 0.9081512659788131,
|
| 8250 |
+
"num_tokens": 3989798.0,
|
| 8251 |
+
"step": 7900
|
| 8252 |
+
},
|
| 8253 |
+
{
|
| 8254 |
+
"entropy": 0.2969894910231233,
|
| 8255 |
+
"epoch": 0.5992537737457149,
|
| 8256 |
+
"grad_norm": 0.5206224322319031,
|
| 8257 |
+
"learning_rate": 3.6538412435318036e-05,
|
| 8258 |
+
"loss": 0.35520272254943847,
|
| 8259 |
+
"mean_token_accuracy": 0.903459957242012,
|
| 8260 |
+
"num_tokens": 4092198.0,
|
| 8261 |
+
"step": 7910
|
| 8262 |
+
},
|
| 8263 |
+
{
|
| 8264 |
+
"entropy": 0.261045260168612,
|
| 8265 |
+
"epoch": 0.6000113638515882,
|
| 8266 |
+
"grad_norm": 0.4172385334968567,
|
| 8267 |
+
"learning_rate": 3.642030286098513e-05,
|
| 8268 |
+
"loss": 0.3094768047332764,
|
| 8269 |
+
"mean_token_accuracy": 0.9193493768572807,
|
| 8270 |
+
"num_tokens": 4192896.0,
|
| 8271 |
+
"step": 7920
|
| 8272 |
+
},
|
| 8273 |
+
{
|
| 8274 |
+
"entropy": 0.3263862133026123,
|
| 8275 |
+
"epoch": 0.6007689539574613,
|
| 8276 |
+
"grad_norm": 0.4942246079444885,
|
| 8277 |
+
"learning_rate": 3.63022750386324e-05,
|
| 8278 |
+
"loss": 0.3887051582336426,
|
| 8279 |
+
"mean_token_accuracy": 0.8954097390174866,
|
| 8280 |
+
"num_tokens": 4286686.0,
|
| 8281 |
+
"step": 7930
|
| 8282 |
+
},
|
| 8283 |
+
{
|
| 8284 |
+
"entropy": 0.2514767138287425,
|
| 8285 |
+
"epoch": 0.6015265440633345,
|
| 8286 |
+
"grad_norm": 0.48457324504852295,
|
| 8287 |
+
"learning_rate": 3.618432967880647e-05,
|
| 8288 |
+
"loss": 0.27435855865478515,
|
| 8289 |
+
"mean_token_accuracy": 0.9226398333907128,
|
| 8290 |
+
"num_tokens": 4370302.0,
|
| 8291 |
+
"step": 7940
|
| 8292 |
+
},
|
| 8293 |
+
{
|
| 8294 |
+
"entropy": 0.3096071735024452,
|
| 8295 |
+
"epoch": 0.6022841341692078,
|
| 8296 |
+
"grad_norm": 0.4468488097190857,
|
| 8297 |
+
"learning_rate": 3.6066467491557554e-05,
|
| 8298 |
+
"loss": 0.36656315326690675,
|
| 8299 |
+
"mean_token_accuracy": 0.9038876727223396,
|
| 8300 |
+
"num_tokens": 4471708.0,
|
| 8301 |
+
"step": 7950
|
| 8302 |
+
},
|
| 8303 |
+
{
|
| 8304 |
+
"entropy": 0.25050507206469774,
|
| 8305 |
+
"epoch": 0.603041724275081,
|
| 8306 |
+
"grad_norm": 0.4183157682418823,
|
| 8307 |
+
"learning_rate": 3.5948689186435143e-05,
|
| 8308 |
+
"loss": 0.2869309425354004,
|
| 8309 |
+
"mean_token_accuracy": 0.9210500806570053,
|
| 8310 |
+
"num_tokens": 4563613.0,
|
| 8311 |
+
"step": 7960
|
| 8312 |
+
},
|
| 8313 |
+
{
|
| 8314 |
+
"entropy": 0.28720583356916907,
|
| 8315 |
+
"epoch": 0.6037993143809541,
|
| 8316 |
+
"grad_norm": 0.45605581998825073,
|
| 8317 |
+
"learning_rate": 3.583099547248373e-05,
|
| 8318 |
+
"loss": 0.32700610160827637,
|
| 8319 |
+
"mean_token_accuracy": 0.9074159458279609,
|
| 8320 |
+
"num_tokens": 4658290.0,
|
| 8321 |
+
"step": 7970
|
| 8322 |
+
},
|
| 8323 |
+
{
|
| 8324 |
+
"entropy": 0.28136809784919026,
|
| 8325 |
+
"epoch": 0.6045569044868274,
|
| 8326 |
+
"grad_norm": 0.5523653030395508,
|
| 8327 |
+
"learning_rate": 3.5713387058238584e-05,
|
| 8328 |
+
"loss": 0.3111879825592041,
|
| 8329 |
+
"mean_token_accuracy": 0.909286692738533,
|
| 8330 |
+
"num_tokens": 4754393.0,
|
| 8331 |
+
"step": 7980
|
| 8332 |
+
},
|
| 8333 |
+
{
|
| 8334 |
+
"entropy": 0.3156980287283659,
|
| 8335 |
+
"epoch": 0.6053144945927006,
|
| 8336 |
+
"grad_norm": 0.4386664628982544,
|
| 8337 |
+
"learning_rate": 3.559586465172143e-05,
|
| 8338 |
+
"loss": 0.3415090322494507,
|
| 8339 |
+
"mean_token_accuracy": 0.9031947538256645,
|
| 8340 |
+
"num_tokens": 4866668.0,
|
| 8341 |
+
"step": 7990
|
| 8342 |
+
},
|
| 8343 |
+
{
|
| 8344 |
+
"entropy": 0.33502144310623405,
|
| 8345 |
+
"epoch": 0.6060720846985739,
|
| 8346 |
+
"grad_norm": 0.4574124217033386,
|
| 8347 |
+
"learning_rate": 3.547842896043624e-05,
|
| 8348 |
+
"loss": 0.3874842405319214,
|
| 8349 |
+
"mean_token_accuracy": 0.8882740035653114,
|
| 8350 |
+
"num_tokens": 4973743.0,
|
| 8351 |
+
"step": 8000
|
| 8352 |
+
},
|
| 8353 |
+
{
|
| 8354 |
+
"epoch": 0.6060720846985739,
|
| 8355 |
+
"eval_entropy": 0.2932357283567221,
|
| 8356 |
+
"eval_loss": 0.2922365963459015,
|
| 8357 |
+
"eval_mean_token_accuracy": 0.9054181920678428,
|
| 8358 |
+
"eval_num_tokens": 4973743.0,
|
| 8359 |
+
"eval_runtime": 365.9511,
|
| 8360 |
+
"eval_samples_per_second": 5.831,
|
| 8361 |
+
"eval_steps_per_second": 1.459,
|
| 8362 |
+
"step": 8000
|
| 8363 |
+
},
|
| 8364 |
+
{
|
| 8365 |
+
"entropy": 0.28289016969501973,
|
| 8366 |
+
"epoch": 0.606829674804447,
|
| 8367 |
+
"grad_norm": 0.4253995418548584,
|
| 8368 |
+
"learning_rate": 3.5361080691364915e-05,
|
| 8369 |
+
"loss": 0.3254977226257324,
|
| 8370 |
+
"mean_token_accuracy": 0.9125159963965416,
|
| 8371 |
+
"num_tokens": 5077217.0,
|
| 8372 |
+
"step": 8010
|
| 8373 |
+
},
|
| 8374 |
+
{
|
| 8375 |
+
"entropy": 0.35271654911339284,
|
| 8376 |
+
"epoch": 0.6075872649103202,
|
| 8377 |
+
"grad_norm": 0.45132774114608765,
|
| 8378 |
+
"learning_rate": 3.524382055096309e-05,
|
| 8379 |
+
"loss": 0.3961679935455322,
|
| 8380 |
+
"mean_token_accuracy": 0.8851734295487403,
|
| 8381 |
+
"num_tokens": 5186347.0,
|
| 8382 |
+
"step": 8020
|
| 8383 |
+
},
|
| 8384 |
+
{
|
| 8385 |
+
"entropy": 0.23756891377270223,
|
| 8386 |
+
"epoch": 0.6083448550161935,
|
| 8387 |
+
"grad_norm": 0.5509276986122131,
|
| 8388 |
+
"learning_rate": 3.5126649245155815e-05,
|
| 8389 |
+
"loss": 0.2758854627609253,
|
| 8390 |
+
"mean_token_accuracy": 0.9254701271653175,
|
| 8391 |
+
"num_tokens": 5278850.0,
|
| 8392 |
+
"step": 8030
|
| 8393 |
+
},
|
| 8394 |
+
{
|
| 8395 |
+
"entropy": 0.29828787613660096,
|
| 8396 |
+
"epoch": 0.6091024451220667,
|
| 8397 |
+
"grad_norm": 0.44331061840057373,
|
| 8398 |
+
"learning_rate": 3.500956747933339e-05,
|
| 8399 |
+
"loss": 0.35719614028930663,
|
| 8400 |
+
"mean_token_accuracy": 0.9053334400057793,
|
| 8401 |
+
"num_tokens": 5375215.0,
|
| 8402 |
+
"step": 8040
|
| 8403 |
+
},
|
| 8404 |
+
{
|
| 8405 |
+
"entropy": 0.2964581159874797,
|
| 8406 |
+
"epoch": 0.60986003522794,
|
| 8407 |
+
"grad_norm": 0.4579179286956787,
|
| 8408 |
+
"learning_rate": 3.4892575958347026e-05,
|
| 8409 |
+
"loss": 0.3475185394287109,
|
| 8410 |
+
"mean_token_accuracy": 0.9046101287007332,
|
| 8411 |
+
"num_tokens": 5475096.0,
|
| 8412 |
+
"step": 8050
|
| 8413 |
+
},
|
| 8414 |
+
{
|
| 8415 |
+
"entropy": 0.2954156167805195,
|
| 8416 |
+
"epoch": 0.6106176253338131,
|
| 8417 |
+
"grad_norm": 0.4848718047142029,
|
| 8418 |
+
"learning_rate": 3.4775675386504656e-05,
|
| 8419 |
+
"loss": 0.3670936346054077,
|
| 8420 |
+
"mean_token_accuracy": 0.9055470645427703,
|
| 8421 |
+
"num_tokens": 5574610.0,
|
| 8422 |
+
"step": 8060
|
| 8423 |
+
},
|
| 8424 |
+
{
|
| 8425 |
+
"entropy": 0.30577495936304333,
|
| 8426 |
+
"epoch": 0.6113752154396863,
|
| 8427 |
+
"grad_norm": 0.4090331494808197,
|
| 8428 |
+
"learning_rate": 3.4658866467566695e-05,
|
| 8429 |
+
"loss": 0.3455164909362793,
|
| 8430 |
+
"mean_token_accuracy": 0.9043252870440484,
|
| 8431 |
+
"num_tokens": 5672698.0,
|
| 8432 |
+
"step": 8070
|
| 8433 |
+
},
|
| 8434 |
+
{
|
| 8435 |
+
"entropy": 0.2794995250180364,
|
| 8436 |
+
"epoch": 0.6121328055455596,
|
| 8437 |
+
"grad_norm": 0.37226206064224243,
|
| 8438 |
+
"learning_rate": 3.45421499047418e-05,
|
| 8439 |
+
"loss": 0.3282626152038574,
|
| 8440 |
+
"mean_token_accuracy": 0.9109877929091453,
|
| 8441 |
+
"num_tokens": 5777564.0,
|
| 8442 |
+
"step": 8080
|
| 8443 |
+
},
|
| 8444 |
+
{
|
| 8445 |
+
"entropy": 0.3204280287027359,
|
| 8446 |
+
"epoch": 0.6128903956514328,
|
| 8447 |
+
"grad_norm": 0.5018744468688965,
|
| 8448 |
+
"learning_rate": 3.442552640068261e-05,
|
| 8449 |
+
"loss": 0.35953049659729003,
|
| 8450 |
+
"mean_token_accuracy": 0.9001020267605782,
|
| 8451 |
+
"num_tokens": 5889298.0,
|
| 8452 |
+
"step": 8090
|
| 8453 |
+
},
|
| 8454 |
+
{
|
| 8455 |
+
"entropy": 0.3048412922769785,
|
| 8456 |
+
"epoch": 0.6136479857573061,
|
| 8457 |
+
"grad_norm": 0.3502022922039032,
|
| 8458 |
+
"learning_rate": 3.430899665748153e-05,
|
| 8459 |
+
"loss": 0.3605944633483887,
|
| 8460 |
+
"mean_token_accuracy": 0.9040819108486176,
|
| 8461 |
+
"num_tokens": 5990146.0,
|
| 8462 |
+
"step": 8100
|
| 8463 |
+
},
|
| 8464 |
+
{
|
| 8465 |
+
"entropy": 0.3116092005744576,
|
| 8466 |
+
"epoch": 0.6144055758631792,
|
| 8467 |
+
"grad_norm": 0.41480910778045654,
|
| 8468 |
+
"learning_rate": 3.419256137666654e-05,
|
| 8469 |
+
"loss": 0.3476060390472412,
|
| 8470 |
+
"mean_token_accuracy": 0.9006553784012794,
|
| 8471 |
+
"num_tokens": 6089746.0,
|
| 8472 |
+
"step": 8110
|
| 8473 |
+
},
|
| 8474 |
+
{
|
| 8475 |
+
"entropy": 0.2927050875499845,
|
| 8476 |
+
"epoch": 0.6151631659690524,
|
| 8477 |
+
"grad_norm": 0.4578903615474701,
|
| 8478 |
+
"learning_rate": 3.4076221259196915e-05,
|
| 8479 |
+
"loss": 0.3382187604904175,
|
| 8480 |
+
"mean_token_accuracy": 0.9089546456933022,
|
| 8481 |
+
"num_tokens": 6189471.0,
|
| 8482 |
+
"step": 8120
|
| 8483 |
+
},
|
| 8484 |
+
{
|
| 8485 |
+
"entropy": 0.2897569188848138,
|
| 8486 |
+
"epoch": 0.6159207560749257,
|
| 8487 |
+
"grad_norm": 0.5105519890785217,
|
| 8488 |
+
"learning_rate": 3.3959977005459036e-05,
|
| 8489 |
+
"loss": 0.36240324974060056,
|
| 8490 |
+
"mean_token_accuracy": 0.9040745571255684,
|
| 8491 |
+
"num_tokens": 6288046.0,
|
| 8492 |
+
"step": 8130
|
| 8493 |
+
},
|
| 8494 |
+
{
|
| 8495 |
+
"entropy": 0.3016464538872242,
|
| 8496 |
+
"epoch": 0.6166783461807989,
|
| 8497 |
+
"grad_norm": 0.4389963448047638,
|
| 8498 |
+
"learning_rate": 3.3843829315262175e-05,
|
| 8499 |
+
"loss": 0.3496075630187988,
|
| 8500 |
+
"mean_token_accuracy": 0.9027613863348961,
|
| 8501 |
+
"num_tokens": 6387427.0,
|
| 8502 |
+
"step": 8140
|
| 8503 |
+
},
|
| 8504 |
+
{
|
| 8505 |
+
"entropy": 0.28618313241750004,
|
| 8506 |
+
"epoch": 0.6174359362866721,
|
| 8507 |
+
"grad_norm": 0.5144838690757751,
|
| 8508 |
+
"learning_rate": 3.372777888783427e-05,
|
| 8509 |
+
"loss": 0.31791436672210693,
|
| 8510 |
+
"mean_token_accuracy": 0.9094356775283814,
|
| 8511 |
+
"num_tokens": 6480239.0,
|
| 8512 |
+
"step": 8150
|
| 8513 |
+
},
|
| 8514 |
+
{
|
| 8515 |
+
"entropy": 0.3418790502473712,
|
| 8516 |
+
"epoch": 0.6181935263925453,
|
| 8517 |
+
"grad_norm": 0.34844839572906494,
|
| 8518 |
+
"learning_rate": 3.361182642181774e-05,
|
| 8519 |
+
"loss": 0.3773730993270874,
|
| 8520 |
+
"mean_token_accuracy": 0.8917641252279281,
|
| 8521 |
+
"num_tokens": 6583332.0,
|
| 8522 |
+
"step": 8160
|
| 8523 |
+
},
|
| 8524 |
+
{
|
| 8525 |
+
"entropy": 0.28220475669950246,
|
| 8526 |
+
"epoch": 0.6189511164984185,
|
| 8527 |
+
"grad_norm": 0.41431310772895813,
|
| 8528 |
+
"learning_rate": 3.349597261526521e-05,
|
| 8529 |
+
"loss": 0.3407491683959961,
|
| 8530 |
+
"mean_token_accuracy": 0.9110078230500221,
|
| 8531 |
+
"num_tokens": 6682980.0,
|
| 8532 |
+
"step": 8170
|
| 8533 |
+
},
|
| 8534 |
+
{
|
| 8535 |
+
"entropy": 0.2900077596306801,
|
| 8536 |
+
"epoch": 0.6197087066042918,
|
| 8537 |
+
"grad_norm": 0.44798147678375244,
|
| 8538 |
+
"learning_rate": 3.338021816563543e-05,
|
| 8539 |
+
"loss": 0.3575183153152466,
|
| 8540 |
+
"mean_token_accuracy": 0.9059248685836792,
|
| 8541 |
+
"num_tokens": 6780719.0,
|
| 8542 |
+
"step": 8180
|
| 8543 |
+
},
|
| 8544 |
+
{
|
| 8545 |
+
"entropy": 0.2807491887360811,
|
| 8546 |
+
"epoch": 0.620466296710165,
|
| 8547 |
+
"grad_norm": 0.3624568283557892,
|
| 8548 |
+
"learning_rate": 3.326456376978895e-05,
|
| 8549 |
+
"loss": 0.31266469955444337,
|
| 8550 |
+
"mean_token_accuracy": 0.9111661002039909,
|
| 8551 |
+
"num_tokens": 6877202.0,
|
| 8552 |
+
"step": 8190
|
| 8553 |
+
},
|
| 8554 |
+
{
|
| 8555 |
+
"entropy": 0.27796862814575435,
|
| 8556 |
+
"epoch": 0.6212238868160381,
|
| 8557 |
+
"grad_norm": 0.4349261522293091,
|
| 8558 |
+
"learning_rate": 3.3149010123984e-05,
|
| 8559 |
+
"loss": 0.33564770221710205,
|
| 8560 |
+
"mean_token_accuracy": 0.912667815387249,
|
| 8561 |
+
"num_tokens": 6974792.0,
|
| 8562 |
+
"step": 8200
|
| 8563 |
+
},
|
| 8564 |
+
{
|
| 8565 |
+
"entropy": 0.3014746461063623,
|
| 8566 |
+
"epoch": 0.6219814769219114,
|
| 8567 |
+
"grad_norm": 0.446551650762558,
|
| 8568 |
+
"learning_rate": 3.303355792387228e-05,
|
| 8569 |
+
"loss": 0.3223909854888916,
|
| 8570 |
+
"mean_token_accuracy": 0.9048597440123558,
|
| 8571 |
+
"num_tokens": 7078884.0,
|
| 8572 |
+
"step": 8210
|
| 8573 |
+
},
|
| 8574 |
+
{
|
| 8575 |
+
"entropy": 0.28495933916419747,
|
| 8576 |
+
"epoch": 0.6227390670277846,
|
| 8577 |
+
"grad_norm": 0.421099454164505,
|
| 8578 |
+
"learning_rate": 3.291820786449477e-05,
|
| 8579 |
+
"loss": 0.33696722984313965,
|
| 8580 |
+
"mean_token_accuracy": 0.9085236057639122,
|
| 8581 |
+
"num_tokens": 7169030.0,
|
| 8582 |
+
"step": 8220
|
| 8583 |
+
},
|
| 8584 |
+
{
|
| 8585 |
+
"entropy": 0.3023871978744864,
|
| 8586 |
+
"epoch": 0.6234966571336579,
|
| 8587 |
+
"grad_norm": 0.3818657100200653,
|
| 8588 |
+
"learning_rate": 3.2802960640277526e-05,
|
| 8589 |
+
"loss": 0.35631389617919923,
|
| 8590 |
+
"mean_token_accuracy": 0.9038525074720383,
|
| 8591 |
+
"num_tokens": 7277118.0,
|
| 8592 |
+
"step": 8230
|
| 8593 |
+
},
|
| 8594 |
+
{
|
| 8595 |
+
"entropy": 0.269647141546011,
|
| 8596 |
+
"epoch": 0.624254247239531,
|
| 8597 |
+
"grad_norm": 0.4466300904750824,
|
| 8598 |
+
"learning_rate": 3.2687816945027564e-05,
|
| 8599 |
+
"loss": 0.3120474100112915,
|
| 8600 |
+
"mean_token_accuracy": 0.9146815970540046,
|
| 8601 |
+
"num_tokens": 7373131.0,
|
| 8602 |
+
"step": 8240
|
| 8603 |
+
},
|
| 8604 |
+
{
|
| 8605 |
+
"entropy": 0.2958039175719023,
|
| 8606 |
+
"epoch": 0.6250118373454042,
|
| 8607 |
+
"grad_norm": 0.48310843110084534,
|
| 8608 |
+
"learning_rate": 3.2572777471928606e-05,
|
| 8609 |
+
"loss": 0.3290383338928223,
|
| 8610 |
+
"mean_token_accuracy": 0.9073579430580139,
|
| 8611 |
+
"num_tokens": 7476714.0,
|
| 8612 |
+
"step": 8250
|
| 8613 |
+
},
|
| 8614 |
+
{
|
| 8615 |
+
"epoch": 0.6250118373454042,
|
| 8616 |
+
"eval_entropy": 0.29268133736560853,
|
| 8617 |
+
"eval_loss": 0.2911636233329773,
|
| 8618 |
+
"eval_mean_token_accuracy": 0.9058593404427003,
|
| 8619 |
+
"eval_num_tokens": 7476714.0,
|
| 8620 |
+
"eval_runtime": 365.863,
|
| 8621 |
+
"eval_samples_per_second": 5.833,
|
| 8622 |
+
"eval_steps_per_second": 1.46,
|
| 8623 |
+
"step": 8250
|
| 8624 |
+
},
|
| 8625 |
+
{
|
| 8626 |
+
"entropy": 0.26329117398709057,
|
| 8627 |
+
"epoch": 0.6257694274512775,
|
| 8628 |
+
"grad_norm": 0.7173141837120056,
|
| 8629 |
+
"learning_rate": 3.245784291353694e-05,
|
| 8630 |
+
"loss": 0.31045899391174314,
|
| 8631 |
+
"mean_token_accuracy": 0.9150152236223221,
|
| 8632 |
+
"num_tokens": 7564673.0,
|
| 8633 |
+
"step": 8260
|
| 8634 |
+
},
|
| 8635 |
+
{
|
| 8636 |
+
"entropy": 0.33611078970134256,
|
| 8637 |
+
"epoch": 0.6265270175571507,
|
| 8638 |
+
"grad_norm": 0.38313695788383484,
|
| 8639 |
+
"learning_rate": 3.234301396177725e-05,
|
| 8640 |
+
"loss": 0.3895453929901123,
|
| 8641 |
+
"mean_token_accuracy": 0.8916040763258934,
|
| 8642 |
+
"num_tokens": 7663473.0,
|
| 8643 |
+
"step": 8270
|
| 8644 |
+
},
|
| 8645 |
+
{
|
| 8646 |
+
"entropy": 0.3024331361055374,
|
| 8647 |
+
"epoch": 0.627284607663024,
|
| 8648 |
+
"grad_norm": 0.4940413534641266,
|
| 8649 |
+
"learning_rate": 3.2228291307938455e-05,
|
| 8650 |
+
"loss": 0.3421007633209229,
|
| 8651 |
+
"mean_token_accuracy": 0.90429967045784,
|
| 8652 |
+
"num_tokens": 7761455.0,
|
| 8653 |
+
"step": 8280
|
| 8654 |
+
},
|
| 8655 |
+
{
|
| 8656 |
+
"entropy": 0.30017769914120435,
|
| 8657 |
+
"epoch": 0.6280421977688971,
|
| 8658 |
+
"grad_norm": 0.5128603577613831,
|
| 8659 |
+
"learning_rate": 3.2113675642669554e-05,
|
| 8660 |
+
"loss": 0.33949527740478513,
|
| 8661 |
+
"mean_token_accuracy": 0.9084667101502418,
|
| 8662 |
+
"num_tokens": 7862353.0,
|
| 8663 |
+
"step": 8290
|
| 8664 |
+
},
|
| 8665 |
+
{
|
| 8666 |
+
"entropy": 0.27928945813328027,
|
| 8667 |
+
"epoch": 0.6287997878747703,
|
| 8668 |
+
"grad_norm": 0.40456345677375793,
|
| 8669 |
+
"learning_rate": 3.199916765597546e-05,
|
| 8670 |
+
"loss": 0.31844272613525393,
|
| 8671 |
+
"mean_token_accuracy": 0.9115223735570908,
|
| 8672 |
+
"num_tokens": 7960022.0,
|
| 8673 |
+
"step": 8300
|
| 8674 |
+
},
|
| 8675 |
+
{
|
| 8676 |
+
"entropy": 0.255427229963243,
|
| 8677 |
+
"epoch": 0.6295573779806436,
|
| 8678 |
+
"grad_norm": 0.4113430976867676,
|
| 8679 |
+
"learning_rate": 3.188476803721281e-05,
|
| 8680 |
+
"loss": 0.2998863935470581,
|
| 8681 |
+
"mean_token_accuracy": 0.9185087248682976,
|
| 8682 |
+
"num_tokens": 8052940.0,
|
| 8683 |
+
"step": 8310
|
| 8684 |
+
},
|
| 8685 |
+
{
|
| 8686 |
+
"entropy": 0.2984560389071703,
|
| 8687 |
+
"epoch": 0.6303149680865168,
|
| 8688 |
+
"grad_norm": 0.5488832592964172,
|
| 8689 |
+
"learning_rate": 3.177047747508588e-05,
|
| 8690 |
+
"loss": 0.3517791748046875,
|
| 8691 |
+
"mean_token_accuracy": 0.9067248597741127,
|
| 8692 |
+
"num_tokens": 8151099.0,
|
| 8693 |
+
"step": 8320
|
| 8694 |
+
},
|
| 8695 |
+
{
|
| 8696 |
+
"entropy": 0.3129190610721707,
|
| 8697 |
+
"epoch": 0.63107255819239,
|
| 8698 |
+
"grad_norm": 0.3750799894332886,
|
| 8699 |
+
"learning_rate": 3.1656296657642404e-05,
|
| 8700 |
+
"loss": 0.33051750659942625,
|
| 8701 |
+
"mean_token_accuracy": 0.9007332503795624,
|
| 8702 |
+
"num_tokens": 8260867.0,
|
| 8703 |
+
"step": 8330
|
| 8704 |
+
},
|
| 8705 |
+
{
|
| 8706 |
+
"entropy": 0.2844857882708311,
|
| 8707 |
+
"epoch": 0.6318301482982632,
|
| 8708 |
+
"grad_norm": 0.5798194408416748,
|
| 8709 |
+
"learning_rate": 3.154222627226943e-05,
|
| 8710 |
+
"loss": 0.33098492622375486,
|
| 8711 |
+
"mean_token_accuracy": 0.9099094271659851,
|
| 8712 |
+
"num_tokens": 8351402.0,
|
| 8713 |
+
"step": 8340
|
| 8714 |
+
},
|
| 8715 |
+
{
|
| 8716 |
+
"entropy": 0.2657761722803116,
|
| 8717 |
+
"epoch": 0.6325877384041364,
|
| 8718 |
+
"grad_norm": 0.4626791775226593,
|
| 8719 |
+
"learning_rate": 3.142826700568918e-05,
|
| 8720 |
+
"loss": 0.31857900619506835,
|
| 8721 |
+
"mean_token_accuracy": 0.9151063993573189,
|
| 8722 |
+
"num_tokens": 8447880.0,
|
| 8723 |
+
"step": 8350
|
| 8724 |
+
},
|
| 8725 |
+
{
|
| 8726 |
+
"entropy": 0.2750039540231228,
|
| 8727 |
+
"epoch": 0.6333453285100097,
|
| 8728 |
+
"grad_norm": 0.34575963020324707,
|
| 8729 |
+
"learning_rate": 3.1314419543954935e-05,
|
| 8730 |
+
"loss": 0.31147184371948244,
|
| 8731 |
+
"mean_token_accuracy": 0.9130986243486404,
|
| 8732 |
+
"num_tokens": 8546379.0,
|
| 8733 |
+
"step": 8360
|
| 8734 |
+
},
|
| 8735 |
+
{
|
| 8736 |
+
"entropy": 0.3098535820841789,
|
| 8737 |
+
"epoch": 0.6341029186158829,
|
| 8738 |
+
"grad_norm": 0.4455592632293701,
|
| 8739 |
+
"learning_rate": 3.120068457244688e-05,
|
| 8740 |
+
"loss": 0.34967617988586425,
|
| 8741 |
+
"mean_token_accuracy": 0.8998353689908981,
|
| 8742 |
+
"num_tokens": 8659921.0,
|
| 8743 |
+
"step": 8370
|
| 8744 |
+
},
|
| 8745 |
+
{
|
| 8746 |
+
"entropy": 0.28903499990701675,
|
| 8747 |
+
"epoch": 0.6348605087217561,
|
| 8748 |
+
"grad_norm": 0.41922202706336975,
|
| 8749 |
+
"learning_rate": 3.1087062775868e-05,
|
| 8750 |
+
"loss": 0.3292752504348755,
|
| 8751 |
+
"mean_token_accuracy": 0.9103768900036812,
|
| 8752 |
+
"num_tokens": 8746266.0,
|
| 8753 |
+
"step": 8380
|
| 8754 |
+
},
|
| 8755 |
+
{
|
| 8756 |
+
"entropy": 0.3025333072990179,
|
| 8757 |
+
"epoch": 0.6356180988276293,
|
| 8758 |
+
"grad_norm": 0.395956814289093,
|
| 8759 |
+
"learning_rate": 3.097355483823995e-05,
|
| 8760 |
+
"loss": 0.37605485916137693,
|
| 8761 |
+
"mean_token_accuracy": 0.9017941072583199,
|
| 8762 |
+
"num_tokens": 8847721.0,
|
| 8763 |
+
"step": 8390
|
| 8764 |
+
},
|
| 8765 |
+
{
|
| 8766 |
+
"entropy": 0.2861843050457537,
|
| 8767 |
+
"epoch": 0.6363756889335025,
|
| 8768 |
+
"grad_norm": 0.41809895634651184,
|
| 8769 |
+
"learning_rate": 3.086016144289893e-05,
|
| 8770 |
+
"loss": 0.33304562568664553,
|
| 8771 |
+
"mean_token_accuracy": 0.9101024970412255,
|
| 8772 |
+
"num_tokens": 8940009.0,
|
| 8773 |
+
"step": 8400
|
| 8774 |
+
},
|
| 8775 |
+
{
|
| 8776 |
+
"entropy": 0.28160879090428353,
|
| 8777 |
+
"epoch": 0.6371332790393758,
|
| 8778 |
+
"grad_norm": 0.367637574672699,
|
| 8779 |
+
"learning_rate": 3.07468832724916e-05,
|
| 8780 |
+
"loss": 0.32176928520202636,
|
| 8781 |
+
"mean_token_accuracy": 0.9087224662303924,
|
| 8782 |
+
"num_tokens": 9043873.0,
|
| 8783 |
+
"step": 8410
|
| 8784 |
+
},
|
| 8785 |
+
{
|
| 8786 |
+
"entropy": 0.3001458944752812,
|
| 8787 |
+
"epoch": 0.637890869145249,
|
| 8788 |
+
"grad_norm": 0.43081459403038025,
|
| 8789 |
+
"learning_rate": 3.063372100897088e-05,
|
| 8790 |
+
"loss": 0.3428102970123291,
|
| 8791 |
+
"mean_token_accuracy": 0.9042011275887489,
|
| 8792 |
+
"num_tokens": 9142258.0,
|
| 8793 |
+
"step": 8420
|
| 8794 |
+
},
|
| 8795 |
+
{
|
| 8796 |
+
"entropy": 0.2861454239115119,
|
| 8797 |
+
"epoch": 0.6386484592511222,
|
| 8798 |
+
"grad_norm": 0.4634046256542206,
|
| 8799 |
+
"learning_rate": 3.052067533359199e-05,
|
| 8800 |
+
"loss": 0.35511579513549807,
|
| 8801 |
+
"mean_token_accuracy": 0.9067476555705071,
|
| 8802 |
+
"num_tokens": 9241331.0,
|
| 8803 |
+
"step": 8430
|
| 8804 |
+
},
|
| 8805 |
+
{
|
| 8806 |
+
"entropy": 0.30580927096307275,
|
| 8807 |
+
"epoch": 0.6394060493569954,
|
| 8808 |
+
"grad_norm": 0.4320875108242035,
|
| 8809 |
+
"learning_rate": 3.0407746926908225e-05,
|
| 8810 |
+
"loss": 0.34864106178283694,
|
| 8811 |
+
"mean_token_accuracy": 0.9023985117673874,
|
| 8812 |
+
"num_tokens": 9337148.0,
|
| 8813 |
+
"step": 8440
|
| 8814 |
+
},
|
| 8815 |
+
{
|
| 8816 |
+
"entropy": 0.25655875001102685,
|
| 8817 |
+
"epoch": 0.6401636394628686,
|
| 8818 |
+
"grad_norm": 0.5615314245223999,
|
| 8819 |
+
"learning_rate": 3.029493646876692e-05,
|
| 8820 |
+
"loss": 0.30922226905822753,
|
| 8821 |
+
"mean_token_accuracy": 0.9156622171401978,
|
| 8822 |
+
"num_tokens": 9431355.0,
|
| 8823 |
+
"step": 8450
|
| 8824 |
+
},
|
| 8825 |
+
{
|
| 8826 |
+
"entropy": 0.255257972329855,
|
| 8827 |
+
"epoch": 0.6409212295687419,
|
| 8828 |
+
"grad_norm": 0.3797270655632019,
|
| 8829 |
+
"learning_rate": 3.0182244638305344e-05,
|
| 8830 |
+
"loss": 0.2795403957366943,
|
| 8831 |
+
"mean_token_accuracy": 0.9213606551289558,
|
| 8832 |
+
"num_tokens": 9514585.0,
|
| 8833 |
+
"step": 8460
|
| 8834 |
+
},
|
| 8835 |
+
{
|
| 8836 |
+
"entropy": 0.28664037846028806,
|
| 8837 |
+
"epoch": 0.641678819674615,
|
| 8838 |
+
"grad_norm": 0.5212521553039551,
|
| 8839 |
+
"learning_rate": 3.0069672113946597e-05,
|
| 8840 |
+
"loss": 0.3244238615036011,
|
| 8841 |
+
"mean_token_accuracy": 0.9093809619545936,
|
| 8842 |
+
"num_tokens": 9610263.0,
|
| 8843 |
+
"step": 8470
|
| 8844 |
+
},
|
| 8845 |
+
{
|
| 8846 |
+
"entropy": 0.30011385809630153,
|
| 8847 |
+
"epoch": 0.6424364097804882,
|
| 8848 |
+
"grad_norm": 0.4266282618045807,
|
| 8849 |
+
"learning_rate": 2.9957219573395535e-05,
|
| 8850 |
+
"loss": 0.3350399971008301,
|
| 8851 |
+
"mean_token_accuracy": 0.9042456343770027,
|
| 8852 |
+
"num_tokens": 9706567.0,
|
| 8853 |
+
"step": 8480
|
| 8854 |
+
},
|
| 8855 |
+
{
|
| 8856 |
+
"entropy": 0.28647223617881534,
|
| 8857 |
+
"epoch": 0.6431939998863615,
|
| 8858 |
+
"grad_norm": 0.4410366117954254,
|
| 8859 |
+
"learning_rate": 2.9844887693634705e-05,
|
| 8860 |
+
"loss": 0.3288688659667969,
|
| 8861 |
+
"mean_token_accuracy": 0.9098197788000106,
|
| 8862 |
+
"num_tokens": 9796673.0,
|
| 8863 |
+
"step": 8490
|
| 8864 |
+
},
|
| 8865 |
+
{
|
| 8866 |
+
"entropy": 0.2846911262720823,
|
| 8867 |
+
"epoch": 0.6439515899922347,
|
| 8868 |
+
"grad_norm": 0.5021436810493469,
|
| 8869 |
+
"learning_rate": 2.9732677150920252e-05,
|
| 8870 |
+
"loss": 0.32644355297088623,
|
| 8871 |
+
"mean_token_accuracy": 0.9078450202941895,
|
| 8872 |
+
"num_tokens": 9893425.0,
|
| 8873 |
+
"step": 8500
|
| 8874 |
+
},
|
| 8875 |
+
{
|
| 8876 |
+
"epoch": 0.6439515899922347,
|
| 8877 |
+
"eval_entropy": 0.29221179915333956,
|
| 8878 |
+
"eval_loss": 0.29068824648857117,
|
| 8879 |
+
"eval_mean_token_accuracy": 0.9061205931817101,
|
| 8880 |
+
"eval_num_tokens": 9893425.0,
|
| 8881 |
+
"eval_runtime": 365.7507,
|
| 8882 |
+
"eval_samples_per_second": 5.835,
|
| 8883 |
+
"eval_steps_per_second": 1.46,
|
| 8884 |
+
"step": 8500
|
| 8885 |
}
|
| 8886 |
],
|
| 8887 |
"logging_steps": 10,
|
|
|
|
| 8901 |
"attributes": {}
|
| 8902 |
}
|
| 8903 |
},
|
| 8904 |
+
"total_flos": 7.38235242401235e+17,
|
| 8905 |
"train_batch_size": 2,
|
| 8906 |
"trial_name": null,
|
| 8907 |
"trial_params": null
|