Training in progress, step 12500, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -30,15 +30,15 @@
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
|
|
|
|
|
|
| 33 |
"w2",
|
|
|
|
| 34 |
"in_proj",
|
| 35 |
-
"q_proj",
|
| 36 |
-
"w1",
|
| 37 |
"k_proj",
|
| 38 |
-
"
|
| 39 |
-
"
|
| 40 |
-
"out_proj"
|
| 41 |
-
"w3"
|
| 42 |
],
|
| 43 |
"target_parameters": null,
|
| 44 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
| 33 |
+
"w3",
|
| 34 |
+
"o_proj",
|
| 35 |
"w2",
|
| 36 |
+
"v_proj",
|
| 37 |
"in_proj",
|
|
|
|
|
|
|
| 38 |
"k_proj",
|
| 39 |
+
"w1",
|
| 40 |
+
"q_proj",
|
| 41 |
+
"out_proj"
|
|
|
|
| 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:f49c1c286799f5e29be9fd64cbb46ba2c512dde244e29ff64142fb3a2518aeae
|
| 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:d403e4711416334edfb83b9e011344ffaef8e568045fb9c0c258a6b9e9117fb0
|
| 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 sha256:52751768493bf41eace4ce1ca4bb02c8a95d8e0ee1948ba13c4a228e4603a1c5
|
| 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:aa95de41fb3d74cdc8a498a9aa118c1a2b4742060ee9fca437abea9b00aab212
|
| 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 sha256:3b1e902a4d420ccb685b217aede2397dc70ed52f0fbd29679fff66c8a6851ff4
|
| 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,
|
|
@@ -11231,6 +11231,1833 @@
|
|
| 11231 |
"eval_samples_per_second": 5.851,
|
| 11232 |
"eval_steps_per_second": 1.464,
|
| 11233 |
"step": 10750
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11234 |
}
|
| 11235 |
],
|
| 11236 |
"logging_steps": 10,
|
|
@@ -11250,7 +13077,7 @@
|
|
| 11250 |
"attributes": {}
|
| 11251 |
}
|
| 11252 |
},
|
| 11253 |
-
"total_flos":
|
| 11254 |
"train_batch_size": 2,
|
| 11255 |
"trial_name": null,
|
| 11256 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 12500,
|
| 3 |
+
"best_metric": 0.2848212718963623,
|
| 4 |
+
"best_model_checkpoint": "./lfm25-kodcode-lora/checkpoint-12500",
|
| 5 |
+
"epoch": 0.9469876323415216,
|
| 6 |
"eval_steps": 250,
|
| 7 |
+
"global_step": 12500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 11231 |
"eval_samples_per_second": 5.851,
|
| 11232 |
"eval_steps_per_second": 1.464,
|
| 11233 |
"step": 10750
|
| 11234 |
+
},
|
| 11235 |
+
{
|
| 11236 |
+
"entropy": 0.2615930700674653,
|
| 11237 |
+
"epoch": 0.8151669539195818,
|
| 11238 |
+
"grad_norm": 0.4948700964450836,
|
| 11239 |
+
"learning_rate": 8.70287588830654e-06,
|
| 11240 |
+
"loss": 0.30023128986358644,
|
| 11241 |
+
"mean_token_accuracy": 0.9172126367688179,
|
| 11242 |
+
"num_tokens": 95394.0,
|
| 11243 |
+
"step": 10760
|
| 11244 |
+
},
|
| 11245 |
+
{
|
| 11246 |
+
"entropy": 0.26154228337109087,
|
| 11247 |
+
"epoch": 0.815924544025455,
|
| 11248 |
+
"grad_norm": 0.3987369239330292,
|
| 11249 |
+
"learning_rate": 8.633838805669543e-06,
|
| 11250 |
+
"loss": 0.3083802223205566,
|
| 11251 |
+
"mean_token_accuracy": 0.916927844285965,
|
| 11252 |
+
"num_tokens": 198965.0,
|
| 11253 |
+
"step": 10770
|
| 11254 |
+
},
|
| 11255 |
+
{
|
| 11256 |
+
"entropy": 0.29399398546665906,
|
| 11257 |
+
"epoch": 0.8166821341313283,
|
| 11258 |
+
"grad_norm": 0.4157302677631378,
|
| 11259 |
+
"learning_rate": 8.565050754034847e-06,
|
| 11260 |
+
"loss": 0.330983567237854,
|
| 11261 |
+
"mean_token_accuracy": 0.907181803882122,
|
| 11262 |
+
"num_tokens": 300936.0,
|
| 11263 |
+
"step": 10780
|
| 11264 |
+
},
|
| 11265 |
+
{
|
| 11266 |
+
"entropy": 0.2811663061380386,
|
| 11267 |
+
"epoch": 0.8174397242372015,
|
| 11268 |
+
"grad_norm": 0.41286197304725647,
|
| 11269 |
+
"learning_rate": 8.496512147517688e-06,
|
| 11270 |
+
"loss": 0.3124821662902832,
|
| 11271 |
+
"mean_token_accuracy": 0.9098848760128021,
|
| 11272 |
+
"num_tokens": 407545.0,
|
| 11273 |
+
"step": 10790
|
| 11274 |
+
},
|
| 11275 |
+
{
|
| 11276 |
+
"entropy": 0.29347006008028986,
|
| 11277 |
+
"epoch": 0.8181973143430746,
|
| 11278 |
+
"grad_norm": 0.4638175666332245,
|
| 11279 |
+
"learning_rate": 8.428223398731588e-06,
|
| 11280 |
+
"loss": 0.32779746055603026,
|
| 11281 |
+
"mean_token_accuracy": 0.9062278017401695,
|
| 11282 |
+
"num_tokens": 511306.0,
|
| 11283 |
+
"step": 10800
|
| 11284 |
+
},
|
| 11285 |
+
{
|
| 11286 |
+
"entropy": 0.3169729994609952,
|
| 11287 |
+
"epoch": 0.8189549044489479,
|
| 11288 |
+
"grad_norm": 0.6172231435775757,
|
| 11289 |
+
"learning_rate": 8.360184918785902e-06,
|
| 11290 |
+
"loss": 0.3641566038131714,
|
| 11291 |
+
"mean_token_accuracy": 0.8966098591685295,
|
| 11292 |
+
"num_tokens": 615936.0,
|
| 11293 |
+
"step": 10810
|
| 11294 |
+
},
|
| 11295 |
+
{
|
| 11296 |
+
"entropy": 0.28694524206221106,
|
| 11297 |
+
"epoch": 0.8197124945548211,
|
| 11298 |
+
"grad_norm": 0.5250155925750732,
|
| 11299 |
+
"learning_rate": 8.292397117283318e-06,
|
| 11300 |
+
"loss": 0.3268028974533081,
|
| 11301 |
+
"mean_token_accuracy": 0.9117160573601722,
|
| 11302 |
+
"num_tokens": 707912.0,
|
| 11303 |
+
"step": 10820
|
| 11304 |
+
},
|
| 11305 |
+
{
|
| 11306 |
+
"entropy": 0.2993797710165381,
|
| 11307 |
+
"epoch": 0.8204700846606944,
|
| 11308 |
+
"grad_norm": 0.450602650642395,
|
| 11309 |
+
"learning_rate": 8.224860402317398e-06,
|
| 11310 |
+
"loss": 0.3751563549041748,
|
| 11311 |
+
"mean_token_accuracy": 0.9021484076976776,
|
| 11312 |
+
"num_tokens": 797927.0,
|
| 11313 |
+
"step": 10830
|
| 11314 |
+
},
|
| 11315 |
+
{
|
| 11316 |
+
"entropy": 0.2887501694262028,
|
| 11317 |
+
"epoch": 0.8212276747665676,
|
| 11318 |
+
"grad_norm": 0.3982142210006714,
|
| 11319 |
+
"learning_rate": 8.157575180470139e-06,
|
| 11320 |
+
"loss": 0.3479593276977539,
|
| 11321 |
+
"mean_token_accuracy": 0.9077468276023865,
|
| 11322 |
+
"num_tokens": 904264.0,
|
| 11323 |
+
"step": 10840
|
| 11324 |
+
},
|
| 11325 |
+
{
|
| 11326 |
+
"entropy": 0.30117298644036056,
|
| 11327 |
+
"epoch": 0.8219852648724407,
|
| 11328 |
+
"grad_norm": 0.4184916913509369,
|
| 11329 |
+
"learning_rate": 8.090541856809465e-06,
|
| 11330 |
+
"loss": 0.33893485069274903,
|
| 11331 |
+
"mean_token_accuracy": 0.9050878256559372,
|
| 11332 |
+
"num_tokens": 1007433.0,
|
| 11333 |
+
"step": 10850
|
| 11334 |
+
},
|
| 11335 |
+
{
|
| 11336 |
+
"entropy": 0.30625202376395466,
|
| 11337 |
+
"epoch": 0.822742854978314,
|
| 11338 |
+
"grad_norm": 0.43937817215919495,
|
| 11339 |
+
"learning_rate": 8.02376083488688e-06,
|
| 11340 |
+
"loss": 0.34802818298339844,
|
| 11341 |
+
"mean_token_accuracy": 0.9029570311307907,
|
| 11342 |
+
"num_tokens": 1106432.0,
|
| 11343 |
+
"step": 10860
|
| 11344 |
+
},
|
| 11345 |
+
{
|
| 11346 |
+
"entropy": 0.29251224398612974,
|
| 11347 |
+
"epoch": 0.8235004450841872,
|
| 11348 |
+
"grad_norm": 0.4894697964191437,
|
| 11349 |
+
"learning_rate": 7.957232516734963e-06,
|
| 11350 |
+
"loss": 0.35279672145843505,
|
| 11351 |
+
"mean_token_accuracy": 0.9066073164343834,
|
| 11352 |
+
"num_tokens": 1204980.0,
|
| 11353 |
+
"step": 10870
|
| 11354 |
+
},
|
| 11355 |
+
{
|
| 11356 |
+
"entropy": 0.3121814550831914,
|
| 11357 |
+
"epoch": 0.8242580351900605,
|
| 11358 |
+
"grad_norm": 0.451427698135376,
|
| 11359 |
+
"learning_rate": 7.890957302865e-06,
|
| 11360 |
+
"loss": 0.37927200794219973,
|
| 11361 |
+
"mean_token_accuracy": 0.8981422513723374,
|
| 11362 |
+
"num_tokens": 1305202.0,
|
| 11363 |
+
"step": 10880
|
| 11364 |
+
},
|
| 11365 |
+
{
|
| 11366 |
+
"entropy": 0.29268699828535316,
|
| 11367 |
+
"epoch": 0.8250156252959336,
|
| 11368 |
+
"grad_norm": 0.3992355465888977,
|
| 11369 |
+
"learning_rate": 7.82493559226451e-06,
|
| 11370 |
+
"loss": 0.3250342607498169,
|
| 11371 |
+
"mean_token_accuracy": 0.9062216132879257,
|
| 11372 |
+
"num_tokens": 1407868.0,
|
| 11373 |
+
"step": 10890
|
| 11374 |
+
},
|
| 11375 |
+
{
|
| 11376 |
+
"entropy": 0.36002040691673753,
|
| 11377 |
+
"epoch": 0.8257732154018068,
|
| 11378 |
+
"grad_norm": 0.4525466561317444,
|
| 11379 |
+
"learning_rate": 7.759167782394916e-06,
|
| 11380 |
+
"loss": 0.4119568824768066,
|
| 11381 |
+
"mean_token_accuracy": 0.8839563131332397,
|
| 11382 |
+
"num_tokens": 1513950.0,
|
| 11383 |
+
"step": 10900
|
| 11384 |
+
},
|
| 11385 |
+
{
|
| 11386 |
+
"entropy": 0.30947884283959864,
|
| 11387 |
+
"epoch": 0.8265308055076801,
|
| 11388 |
+
"grad_norm": 0.4585402309894562,
|
| 11389 |
+
"learning_rate": 7.69365426918911e-06,
|
| 11390 |
+
"loss": 0.3610663414001465,
|
| 11391 |
+
"mean_token_accuracy": 0.9016437903046608,
|
| 11392 |
+
"num_tokens": 1613270.0,
|
| 11393 |
+
"step": 10910
|
| 11394 |
+
},
|
| 11395 |
+
{
|
| 11396 |
+
"entropy": 0.29806773774325845,
|
| 11397 |
+
"epoch": 0.8272883956135533,
|
| 11398 |
+
"grad_norm": 0.43951550126075745,
|
| 11399 |
+
"learning_rate": 7.628395447049058e-06,
|
| 11400 |
+
"loss": 0.34964966773986816,
|
| 11401 |
+
"mean_token_accuracy": 0.9043863728642464,
|
| 11402 |
+
"num_tokens": 1710564.0,
|
| 11403 |
+
"step": 10920
|
| 11404 |
+
},
|
| 11405 |
+
{
|
| 11406 |
+
"entropy": 0.27706831842660906,
|
| 11407 |
+
"epoch": 0.8280459857194266,
|
| 11408 |
+
"grad_norm": 0.43634042143821716,
|
| 11409 |
+
"learning_rate": 7.563391708843465e-06,
|
| 11410 |
+
"loss": 0.31148567199707033,
|
| 11411 |
+
"mean_token_accuracy": 0.9122755557298661,
|
| 11412 |
+
"num_tokens": 1804996.0,
|
| 11413 |
+
"step": 10930
|
| 11414 |
+
},
|
| 11415 |
+
{
|
| 11416 |
+
"entropy": 0.33176610227674247,
|
| 11417 |
+
"epoch": 0.8288035758252997,
|
| 11418 |
+
"grad_norm": 0.39294061064720154,
|
| 11419 |
+
"learning_rate": 7.498643445905385e-06,
|
| 11420 |
+
"loss": 0.37499704360961916,
|
| 11421 |
+
"mean_token_accuracy": 0.8953326553106308,
|
| 11422 |
+
"num_tokens": 1914019.0,
|
| 11423 |
+
"step": 10940
|
| 11424 |
+
},
|
| 11425 |
+
{
|
| 11426 |
+
"entropy": 0.2956154864281416,
|
| 11427 |
+
"epoch": 0.8295611659311729,
|
| 11428 |
+
"grad_norm": 0.4650717079639435,
|
| 11429 |
+
"learning_rate": 7.434151048029858e-06,
|
| 11430 |
+
"loss": 0.3409877777099609,
|
| 11431 |
+
"mean_token_accuracy": 0.9076192870736122,
|
| 11432 |
+
"num_tokens": 2009215.0,
|
| 11433 |
+
"step": 10950
|
| 11434 |
+
},
|
| 11435 |
+
{
|
| 11436 |
+
"entropy": 0.30589213948696853,
|
| 11437 |
+
"epoch": 0.8303187560370462,
|
| 11438 |
+
"grad_norm": 0.422702819108963,
|
| 11439 |
+
"learning_rate": 7.369914903471592e-06,
|
| 11440 |
+
"loss": 0.3454046487808228,
|
| 11441 |
+
"mean_token_accuracy": 0.9033974334597588,
|
| 11442 |
+
"num_tokens": 2113318.0,
|
| 11443 |
+
"step": 10960
|
| 11444 |
+
},
|
| 11445 |
+
{
|
| 11446 |
+
"entropy": 0.2733181806281209,
|
| 11447 |
+
"epoch": 0.8310763461429194,
|
| 11448 |
+
"grad_norm": 0.47341540455818176,
|
| 11449 |
+
"learning_rate": 7.305935398942598e-06,
|
| 11450 |
+
"loss": 0.3034660816192627,
|
| 11451 |
+
"mean_token_accuracy": 0.9125015497207641,
|
| 11452 |
+
"num_tokens": 2212212.0,
|
| 11453 |
+
"step": 10970
|
| 11454 |
+
},
|
| 11455 |
+
{
|
| 11456 |
+
"entropy": 0.32005896270275114,
|
| 11457 |
+
"epoch": 0.8318339362487925,
|
| 11458 |
+
"grad_norm": 0.4640633761882782,
|
| 11459 |
+
"learning_rate": 7.2422129196098695e-06,
|
| 11460 |
+
"loss": 0.3682070255279541,
|
| 11461 |
+
"mean_token_accuracy": 0.8993272170424461,
|
| 11462 |
+
"num_tokens": 2309913.0,
|
| 11463 |
+
"step": 10980
|
| 11464 |
+
},
|
| 11465 |
+
{
|
| 11466 |
+
"entropy": 0.29598398655653,
|
| 11467 |
+
"epoch": 0.8325915263546658,
|
| 11468 |
+
"grad_norm": 0.43602728843688965,
|
| 11469 |
+
"learning_rate": 7.178747849093093e-06,
|
| 11470 |
+
"loss": 0.34798197746276854,
|
| 11471 |
+
"mean_token_accuracy": 0.9048758611083031,
|
| 11472 |
+
"num_tokens": 2406011.0,
|
| 11473 |
+
"step": 10990
|
| 11474 |
+
},
|
| 11475 |
+
{
|
| 11476 |
+
"entropy": 0.2882571941241622,
|
| 11477 |
+
"epoch": 0.833349116460539,
|
| 11478 |
+
"grad_norm": 0.3429960608482361,
|
| 11479 |
+
"learning_rate": 7.115540569462254e-06,
|
| 11480 |
+
"loss": 0.31615209579467773,
|
| 11481 |
+
"mean_token_accuracy": 0.9083688676357269,
|
| 11482 |
+
"num_tokens": 2498979.0,
|
| 11483 |
+
"step": 11000
|
| 11484 |
+
},
|
| 11485 |
+
{
|
| 11486 |
+
"epoch": 0.833349116460539,
|
| 11487 |
+
"eval_entropy": 0.2886375365185827,
|
| 11488 |
+
"eval_loss": 0.28569018840789795,
|
| 11489 |
+
"eval_mean_token_accuracy": 0.9073382049017631,
|
| 11490 |
+
"eval_num_tokens": 2498979.0,
|
| 11491 |
+
"eval_runtime": 363.4037,
|
| 11492 |
+
"eval_samples_per_second": 5.872,
|
| 11493 |
+
"eval_steps_per_second": 1.469,
|
| 11494 |
+
"step": 11000
|
| 11495 |
+
},
|
| 11496 |
+
{
|
| 11497 |
+
"entropy": 0.32097633071243764,
|
| 11498 |
+
"epoch": 0.8341067065664123,
|
| 11499 |
+
"grad_norm": 0.4701482057571411,
|
| 11500 |
+
"learning_rate": 7.052591461235453e-06,
|
| 11501 |
+
"loss": 0.37491984367370607,
|
| 11502 |
+
"mean_token_accuracy": 0.8957624956965446,
|
| 11503 |
+
"num_tokens": 2599721.0,
|
| 11504 |
+
"step": 11010
|
| 11505 |
+
},
|
| 11506 |
+
{
|
| 11507 |
+
"entropy": 0.3161700405180454,
|
| 11508 |
+
"epoch": 0.8348642966722855,
|
| 11509 |
+
"grad_norm": 0.37062007188796997,
|
| 11510 |
+
"learning_rate": 6.989900903376534e-06,
|
| 11511 |
+
"loss": 0.37365403175354006,
|
| 11512 |
+
"mean_token_accuracy": 0.8983507692813874,
|
| 11513 |
+
"num_tokens": 2704511.0,
|
| 11514 |
+
"step": 11020
|
| 11515 |
+
},
|
| 11516 |
+
{
|
| 11517 |
+
"entropy": 0.3192948242649436,
|
| 11518 |
+
"epoch": 0.8356218867781586,
|
| 11519 |
+
"grad_norm": 0.4017787277698517,
|
| 11520 |
+
"learning_rate": 6.927469273292819e-06,
|
| 11521 |
+
"loss": 0.3574330806732178,
|
| 11522 |
+
"mean_token_accuracy": 0.8968805193901062,
|
| 11523 |
+
"num_tokens": 2810832.0,
|
| 11524 |
+
"step": 11030
|
| 11525 |
+
},
|
| 11526 |
+
{
|
| 11527 |
+
"entropy": 0.3310425169765949,
|
| 11528 |
+
"epoch": 0.8363794768840319,
|
| 11529 |
+
"grad_norm": 0.466722309589386,
|
| 11530 |
+
"learning_rate": 6.865296946832866e-06,
|
| 11531 |
+
"loss": 0.3679776668548584,
|
| 11532 |
+
"mean_token_accuracy": 0.8972247004508972,
|
| 11533 |
+
"num_tokens": 2913394.0,
|
| 11534 |
+
"step": 11040
|
| 11535 |
+
},
|
| 11536 |
+
{
|
| 11537 |
+
"entropy": 0.3021887818351388,
|
| 11538 |
+
"epoch": 0.8371370669899051,
|
| 11539 |
+
"grad_norm": 0.33357468247413635,
|
| 11540 |
+
"learning_rate": 6.8033842982841545e-06,
|
| 11541 |
+
"loss": 0.3556036949157715,
|
| 11542 |
+
"mean_token_accuracy": 0.9023103356361389,
|
| 11543 |
+
"num_tokens": 3011528.0,
|
| 11544 |
+
"step": 11050
|
| 11545 |
+
},
|
| 11546 |
+
{
|
| 11547 |
+
"entropy": 0.3188148533925414,
|
| 11548 |
+
"epoch": 0.8378946570957784,
|
| 11549 |
+
"grad_norm": 0.5248581767082214,
|
| 11550 |
+
"learning_rate": 6.741731700370885e-06,
|
| 11551 |
+
"loss": 0.3591402530670166,
|
| 11552 |
+
"mean_token_accuracy": 0.8997363194823265,
|
| 11553 |
+
"num_tokens": 3114520.0,
|
| 11554 |
+
"step": 11060
|
| 11555 |
+
},
|
| 11556 |
+
{
|
| 11557 |
+
"entropy": 0.28566175922751424,
|
| 11558 |
+
"epoch": 0.8386522472016515,
|
| 11559 |
+
"grad_norm": 0.4290429651737213,
|
| 11560 |
+
"learning_rate": 6.680339524251694e-06,
|
| 11561 |
+
"loss": 0.33682057857513426,
|
| 11562 |
+
"mean_token_accuracy": 0.9096178874373436,
|
| 11563 |
+
"num_tokens": 3220100.0,
|
| 11564 |
+
"step": 11070
|
| 11565 |
+
},
|
| 11566 |
+
{
|
| 11567 |
+
"entropy": 0.2704159118235111,
|
| 11568 |
+
"epoch": 0.8394098373075247,
|
| 11569 |
+
"grad_norm": 0.5517525672912598,
|
| 11570 |
+
"learning_rate": 6.619208139517441e-06,
|
| 11571 |
+
"loss": 0.3258359909057617,
|
| 11572 |
+
"mean_token_accuracy": 0.9156906500458717,
|
| 11573 |
+
"num_tokens": 3311139.0,
|
| 11574 |
+
"step": 11080
|
| 11575 |
+
},
|
| 11576 |
+
{
|
| 11577 |
+
"entropy": 0.2802377393469214,
|
| 11578 |
+
"epoch": 0.840167427413398,
|
| 11579 |
+
"grad_norm": 0.43798232078552246,
|
| 11580 |
+
"learning_rate": 6.558337914188989e-06,
|
| 11581 |
+
"loss": 0.31688680648803713,
|
| 11582 |
+
"mean_token_accuracy": 0.9112754926085472,
|
| 11583 |
+
"num_tokens": 3408982.0,
|
| 11584 |
+
"step": 11090
|
| 11585 |
+
},
|
| 11586 |
+
{
|
| 11587 |
+
"entropy": 0.2785337887704372,
|
| 11588 |
+
"epoch": 0.8409250175192712,
|
| 11589 |
+
"grad_norm": 0.48913341760635376,
|
| 11590 |
+
"learning_rate": 6.497729214714959e-06,
|
| 11591 |
+
"loss": 0.33413968086242674,
|
| 11592 |
+
"mean_token_accuracy": 0.9114461779594422,
|
| 11593 |
+
"num_tokens": 3506297.0,
|
| 11594 |
+
"step": 11100
|
| 11595 |
+
},
|
| 11596 |
+
{
|
| 11597 |
+
"entropy": 0.25448127966374157,
|
| 11598 |
+
"epoch": 0.8416826076251445,
|
| 11599 |
+
"grad_norm": 0.6489408612251282,
|
| 11600 |
+
"learning_rate": 6.437382405969572e-06,
|
| 11601 |
+
"loss": 0.3060408115386963,
|
| 11602 |
+
"mean_token_accuracy": 0.9170120373368263,
|
| 11603 |
+
"num_tokens": 3599883.0,
|
| 11604 |
+
"step": 11110
|
| 11605 |
+
},
|
| 11606 |
+
{
|
| 11607 |
+
"entropy": 0.2931193061172962,
|
| 11608 |
+
"epoch": 0.8424401977310176,
|
| 11609 |
+
"grad_norm": 0.39941656589508057,
|
| 11610 |
+
"learning_rate": 6.377297851250396e-06,
|
| 11611 |
+
"loss": 0.34188551902770997,
|
| 11612 |
+
"mean_token_accuracy": 0.9045077741146088,
|
| 11613 |
+
"num_tokens": 3695770.0,
|
| 11614 |
+
"step": 11120
|
| 11615 |
+
},
|
| 11616 |
+
{
|
| 11617 |
+
"entropy": 0.28530145399272444,
|
| 11618 |
+
"epoch": 0.8431977878368908,
|
| 11619 |
+
"grad_norm": 0.5832393169403076,
|
| 11620 |
+
"learning_rate": 6.317475912276205e-06,
|
| 11621 |
+
"loss": 0.3301248073577881,
|
| 11622 |
+
"mean_token_accuracy": 0.9066288366913795,
|
| 11623 |
+
"num_tokens": 3792849.0,
|
| 11624 |
+
"step": 11130
|
| 11625 |
+
},
|
| 11626 |
+
{
|
| 11627 |
+
"entropy": 0.28422479014843705,
|
| 11628 |
+
"epoch": 0.8439553779427641,
|
| 11629 |
+
"grad_norm": 0.5767186284065247,
|
| 11630 |
+
"learning_rate": 6.257916949184772e-06,
|
| 11631 |
+
"loss": 0.34296748638153074,
|
| 11632 |
+
"mean_token_accuracy": 0.9086409032344818,
|
| 11633 |
+
"num_tokens": 3886352.0,
|
| 11634 |
+
"step": 11140
|
| 11635 |
+
},
|
| 11636 |
+
{
|
| 11637 |
+
"entropy": 0.33191717118024827,
|
| 11638 |
+
"epoch": 0.8447129680486373,
|
| 11639 |
+
"grad_norm": 0.31320858001708984,
|
| 11640 |
+
"learning_rate": 6.1986213205307285e-06,
|
| 11641 |
+
"loss": 0.38002519607543944,
|
| 11642 |
+
"mean_token_accuracy": 0.8913838744163514,
|
| 11643 |
+
"num_tokens": 3990242.0,
|
| 11644 |
+
"step": 11150
|
| 11645 |
+
},
|
| 11646 |
+
{
|
| 11647 |
+
"entropy": 0.27569183316081763,
|
| 11648 |
+
"epoch": 0.8454705581545106,
|
| 11649 |
+
"grad_norm": 0.35137414932250977,
|
| 11650 |
+
"learning_rate": 6.139589383283367e-06,
|
| 11651 |
+
"loss": 0.3208441734313965,
|
| 11652 |
+
"mean_token_accuracy": 0.9123941317200661,
|
| 11653 |
+
"num_tokens": 4090031.0,
|
| 11654 |
+
"step": 11160
|
| 11655 |
+
},
|
| 11656 |
+
{
|
| 11657 |
+
"entropy": 0.26733697205781937,
|
| 11658 |
+
"epoch": 0.8462281482603837,
|
| 11659 |
+
"grad_norm": 0.4320410192012787,
|
| 11660 |
+
"learning_rate": 6.08082149282454e-06,
|
| 11661 |
+
"loss": 0.3125286102294922,
|
| 11662 |
+
"mean_token_accuracy": 0.914748401939869,
|
| 11663 |
+
"num_tokens": 4185773.0,
|
| 11664 |
+
"step": 11170
|
| 11665 |
+
},
|
| 11666 |
+
{
|
| 11667 |
+
"entropy": 0.23407292291522025,
|
| 11668 |
+
"epoch": 0.8469857383662569,
|
| 11669 |
+
"grad_norm": 0.48707106709480286,
|
| 11670 |
+
"learning_rate": 6.022318002946492e-06,
|
| 11671 |
+
"loss": 0.2711918830871582,
|
| 11672 |
+
"mean_token_accuracy": 0.9267101764678956,
|
| 11673 |
+
"num_tokens": 4278787.0,
|
| 11674 |
+
"step": 11180
|
| 11675 |
+
},
|
| 11676 |
+
{
|
| 11677 |
+
"entropy": 0.316350325755775,
|
| 11678 |
+
"epoch": 0.8477433284721302,
|
| 11679 |
+
"grad_norm": 0.3188752830028534,
|
| 11680 |
+
"learning_rate": 5.964079265849698e-06,
|
| 11681 |
+
"loss": 0.3772554397583008,
|
| 11682 |
+
"mean_token_accuracy": 0.8990112885832786,
|
| 11683 |
+
"num_tokens": 4385286.0,
|
| 11684 |
+
"step": 11190
|
| 11685 |
+
},
|
| 11686 |
+
{
|
| 11687 |
+
"entropy": 0.28382357601076363,
|
| 11688 |
+
"epoch": 0.8485009185780034,
|
| 11689 |
+
"grad_norm": 0.5237545371055603,
|
| 11690 |
+
"learning_rate": 5.906105632140818e-06,
|
| 11691 |
+
"loss": 0.33560957908630373,
|
| 11692 |
+
"mean_token_accuracy": 0.9108323022723198,
|
| 11693 |
+
"num_tokens": 4478235.0,
|
| 11694 |
+
"step": 11200
|
| 11695 |
+
},
|
| 11696 |
+
{
|
| 11697 |
+
"entropy": 0.29501939918845893,
|
| 11698 |
+
"epoch": 0.8492585086838765,
|
| 11699 |
+
"grad_norm": 0.4553075134754181,
|
| 11700 |
+
"learning_rate": 5.848397450830539e-06,
|
| 11701 |
+
"loss": 0.3474492073059082,
|
| 11702 |
+
"mean_token_accuracy": 0.9073790535330772,
|
| 11703 |
+
"num_tokens": 4574842.0,
|
| 11704 |
+
"step": 11210
|
| 11705 |
+
},
|
| 11706 |
+
{
|
| 11707 |
+
"entropy": 0.2461070492863655,
|
| 11708 |
+
"epoch": 0.8500160987897498,
|
| 11709 |
+
"grad_norm": 0.3291778564453125,
|
| 11710 |
+
"learning_rate": 5.7909550693314665e-06,
|
| 11711 |
+
"loss": 0.2918540477752686,
|
| 11712 |
+
"mean_token_accuracy": 0.9265199154615402,
|
| 11713 |
+
"num_tokens": 4661233.0,
|
| 11714 |
+
"step": 11220
|
| 11715 |
+
},
|
| 11716 |
+
{
|
| 11717 |
+
"entropy": 0.2952361168339849,
|
| 11718 |
+
"epoch": 0.850773688895623,
|
| 11719 |
+
"grad_norm": 0.5359086990356445,
|
| 11720 |
+
"learning_rate": 5.73377883345606e-06,
|
| 11721 |
+
"loss": 0.3461754560470581,
|
| 11722 |
+
"mean_token_accuracy": 0.9084222391247749,
|
| 11723 |
+
"num_tokens": 4753367.0,
|
| 11724 |
+
"step": 11230
|
| 11725 |
+
},
|
| 11726 |
+
{
|
| 11727 |
+
"entropy": 0.28638926018029454,
|
| 11728 |
+
"epoch": 0.8515312790014963,
|
| 11729 |
+
"grad_norm": 0.4929027855396271,
|
| 11730 |
+
"learning_rate": 5.676869087414544e-06,
|
| 11731 |
+
"loss": 0.3351025104522705,
|
| 11732 |
+
"mean_token_accuracy": 0.9095506638288497,
|
| 11733 |
+
"num_tokens": 4846891.0,
|
| 11734 |
+
"step": 11240
|
| 11735 |
+
},
|
| 11736 |
+
{
|
| 11737 |
+
"entropy": 0.2919112669304013,
|
| 11738 |
+
"epoch": 0.8522888691073695,
|
| 11739 |
+
"grad_norm": 0.43144306540489197,
|
| 11740 |
+
"learning_rate": 5.6202261738128145e-06,
|
| 11741 |
+
"loss": 0.3436851739883423,
|
| 11742 |
+
"mean_token_accuracy": 0.908247809112072,
|
| 11743 |
+
"num_tokens": 4945534.0,
|
| 11744 |
+
"step": 11250
|
| 11745 |
+
},
|
| 11746 |
+
{
|
| 11747 |
+
"epoch": 0.8522888691073695,
|
| 11748 |
+
"eval_entropy": 0.28951325608689465,
|
| 11749 |
+
"eval_loss": 0.28531700372695923,
|
| 11750 |
+
"eval_mean_token_accuracy": 0.907498625333836,
|
| 11751 |
+
"eval_num_tokens": 4945534.0,
|
| 11752 |
+
"eval_runtime": 363.816,
|
| 11753 |
+
"eval_samples_per_second": 5.866,
|
| 11754 |
+
"eval_steps_per_second": 1.468,
|
| 11755 |
+
"step": 11250
|
| 11756 |
+
},
|
| 11757 |
+
{
|
| 11758 |
+
"entropy": 0.3246528644114733,
|
| 11759 |
+
"epoch": 0.8530464592132426,
|
| 11760 |
+
"grad_norm": 0.5182899832725525,
|
| 11761 |
+
"learning_rate": 5.563850433650408e-06,
|
| 11762 |
+
"loss": 0.3830410003662109,
|
| 11763 |
+
"mean_token_accuracy": 0.8927986115217209,
|
| 11764 |
+
"num_tokens": 5053848.0,
|
| 11765 |
+
"step": 11260
|
| 11766 |
+
},
|
| 11767 |
+
{
|
| 11768 |
+
"entropy": 0.2774768492206931,
|
| 11769 |
+
"epoch": 0.8538040493191159,
|
| 11770 |
+
"grad_norm": 0.45578455924987793,
|
| 11771 |
+
"learning_rate": 5.507742206318422e-06,
|
| 11772 |
+
"loss": 0.3247124195098877,
|
| 11773 |
+
"mean_token_accuracy": 0.9107116803526878,
|
| 11774 |
+
"num_tokens": 5150160.0,
|
| 11775 |
+
"step": 11270
|
| 11776 |
+
},
|
| 11777 |
+
{
|
| 11778 |
+
"entropy": 0.2593419700860977,
|
| 11779 |
+
"epoch": 0.8545616394249891,
|
| 11780 |
+
"grad_norm": 0.4656168818473816,
|
| 11781 |
+
"learning_rate": 5.4519018295974945e-06,
|
| 11782 |
+
"loss": 0.29569401741027834,
|
| 11783 |
+
"mean_token_accuracy": 0.9186507374048233,
|
| 11784 |
+
"num_tokens": 5245280.0,
|
| 11785 |
+
"step": 11280
|
| 11786 |
+
},
|
| 11787 |
+
{
|
| 11788 |
+
"entropy": 0.2584766630083323,
|
| 11789 |
+
"epoch": 0.8553192295308624,
|
| 11790 |
+
"grad_norm": 0.36138275265693665,
|
| 11791 |
+
"learning_rate": 5.396329639655756e-06,
|
| 11792 |
+
"loss": 0.2868622064590454,
|
| 11793 |
+
"mean_token_accuracy": 0.9181327551603318,
|
| 11794 |
+
"num_tokens": 5345185.0,
|
| 11795 |
+
"step": 11290
|
| 11796 |
+
},
|
| 11797 |
+
{
|
| 11798 |
+
"entropy": 0.29062068052589896,
|
| 11799 |
+
"epoch": 0.8560768196367355,
|
| 11800 |
+
"grad_norm": 0.3519483804702759,
|
| 11801 |
+
"learning_rate": 5.341025971046798e-06,
|
| 11802 |
+
"loss": 0.3423572540283203,
|
| 11803 |
+
"mean_token_accuracy": 0.9081257462501526,
|
| 11804 |
+
"num_tokens": 5445289.0,
|
| 11805 |
+
"step": 11300
|
| 11806 |
+
},
|
| 11807 |
+
{
|
| 11808 |
+
"entropy": 0.27519712056964635,
|
| 11809 |
+
"epoch": 0.8568344097426087,
|
| 11810 |
+
"grad_norm": 0.3887447416782379,
|
| 11811 |
+
"learning_rate": 5.285991156707681e-06,
|
| 11812 |
+
"loss": 0.317475700378418,
|
| 11813 |
+
"mean_token_accuracy": 0.9138319388031959,
|
| 11814 |
+
"num_tokens": 5544079.0,
|
| 11815 |
+
"step": 11310
|
| 11816 |
+
},
|
| 11817 |
+
{
|
| 11818 |
+
"entropy": 0.2655044592916965,
|
| 11819 |
+
"epoch": 0.857591999848482,
|
| 11820 |
+
"grad_norm": 0.5422892570495605,
|
| 11821 |
+
"learning_rate": 5.2312255279569236e-06,
|
| 11822 |
+
"loss": 0.3148800373077393,
|
| 11823 |
+
"mean_token_accuracy": 0.9163942500948906,
|
| 11824 |
+
"num_tokens": 5636934.0,
|
| 11825 |
+
"step": 11320
|
| 11826 |
+
},
|
| 11827 |
+
{
|
| 11828 |
+
"entropy": 0.26932154782116413,
|
| 11829 |
+
"epoch": 0.8583495899543552,
|
| 11830 |
+
"grad_norm": 0.3854497969150543,
|
| 11831 |
+
"learning_rate": 5.176729414492493e-06,
|
| 11832 |
+
"loss": 0.32476861476898194,
|
| 11833 |
+
"mean_token_accuracy": 0.9170946836471557,
|
| 11834 |
+
"num_tokens": 5734349.0,
|
| 11835 |
+
"step": 11330
|
| 11836 |
+
},
|
| 11837 |
+
{
|
| 11838 |
+
"entropy": 0.2900954818353057,
|
| 11839 |
+
"epoch": 0.8591071800602285,
|
| 11840 |
+
"grad_norm": 0.40638411045074463,
|
| 11841 |
+
"learning_rate": 5.122503144389834e-06,
|
| 11842 |
+
"loss": 0.33924498558044436,
|
| 11843 |
+
"mean_token_accuracy": 0.9079746246337891,
|
| 11844 |
+
"num_tokens": 5835640.0,
|
| 11845 |
+
"step": 11340
|
| 11846 |
+
},
|
| 11847 |
+
{
|
| 11848 |
+
"entropy": 0.3010376073420048,
|
| 11849 |
+
"epoch": 0.8598647701661016,
|
| 11850 |
+
"grad_norm": 0.44111737608909607,
|
| 11851 |
+
"learning_rate": 5.068547044099891e-06,
|
| 11852 |
+
"loss": 0.31704282760620117,
|
| 11853 |
+
"mean_token_accuracy": 0.9046201959252358,
|
| 11854 |
+
"num_tokens": 5934993.0,
|
| 11855 |
+
"step": 11350
|
| 11856 |
+
},
|
| 11857 |
+
{
|
| 11858 |
+
"entropy": 0.3023892106488347,
|
| 11859 |
+
"epoch": 0.8606223602719748,
|
| 11860 |
+
"grad_norm": 0.37891948223114014,
|
| 11861 |
+
"learning_rate": 5.014861438447144e-06,
|
| 11862 |
+
"loss": 0.3835698366165161,
|
| 11863 |
+
"mean_token_accuracy": 0.9033500358462334,
|
| 11864 |
+
"num_tokens": 6038266.0,
|
| 11865 |
+
"step": 11360
|
| 11866 |
+
},
|
| 11867 |
+
{
|
| 11868 |
+
"entropy": 0.2630012549459934,
|
| 11869 |
+
"epoch": 0.8613799503778481,
|
| 11870 |
+
"grad_norm": 0.38084009289741516,
|
| 11871 |
+
"learning_rate": 4.961446650627644e-06,
|
| 11872 |
+
"loss": 0.30712451934814455,
|
| 11873 |
+
"mean_token_accuracy": 0.917777256667614,
|
| 11874 |
+
"num_tokens": 6141215.0,
|
| 11875 |
+
"step": 11370
|
| 11876 |
+
},
|
| 11877 |
+
{
|
| 11878 |
+
"entropy": 0.30645178388804195,
|
| 11879 |
+
"epoch": 0.8621375404837213,
|
| 11880 |
+
"grad_norm": 0.4121268093585968,
|
| 11881 |
+
"learning_rate": 4.9083030022070985e-06,
|
| 11882 |
+
"loss": 0.34171285629272463,
|
| 11883 |
+
"mean_token_accuracy": 0.9038849920034409,
|
| 11884 |
+
"num_tokens": 6235571.0,
|
| 11885 |
+
"step": 11380
|
| 11886 |
+
},
|
| 11887 |
+
{
|
| 11888 |
+
"entropy": 0.2727683449164033,
|
| 11889 |
+
"epoch": 0.8628951305895946,
|
| 11890 |
+
"grad_norm": 0.364166796207428,
|
| 11891 |
+
"learning_rate": 4.855430813118872e-06,
|
| 11892 |
+
"loss": 0.31318480968475343,
|
| 11893 |
+
"mean_token_accuracy": 0.915680679678917,
|
| 11894 |
+
"num_tokens": 6333413.0,
|
| 11895 |
+
"step": 11390
|
| 11896 |
+
},
|
| 11897 |
+
{
|
| 11898 |
+
"entropy": 0.3058678261935711,
|
| 11899 |
+
"epoch": 0.8636527206954677,
|
| 11900 |
+
"grad_norm": 0.37689122557640076,
|
| 11901 |
+
"learning_rate": 4.802830401662129e-06,
|
| 11902 |
+
"loss": 0.3599482297897339,
|
| 11903 |
+
"mean_token_accuracy": 0.9037078693509102,
|
| 11904 |
+
"num_tokens": 6435643.0,
|
| 11905 |
+
"step": 11400
|
| 11906 |
+
},
|
| 11907 |
+
{
|
| 11908 |
+
"entropy": 0.28505960423499344,
|
| 11909 |
+
"epoch": 0.8644103108013409,
|
| 11910 |
+
"grad_norm": 0.6746123433113098,
|
| 11911 |
+
"learning_rate": 4.750502084499886e-06,
|
| 11912 |
+
"loss": 0.3244886875152588,
|
| 11913 |
+
"mean_token_accuracy": 0.9097675696015358,
|
| 11914 |
+
"num_tokens": 6534257.0,
|
| 11915 |
+
"step": 11410
|
| 11916 |
+
},
|
| 11917 |
+
{
|
| 11918 |
+
"entropy": 0.3098301423713565,
|
| 11919 |
+
"epoch": 0.8651679009072142,
|
| 11920 |
+
"grad_norm": 0.500696063041687,
|
| 11921 |
+
"learning_rate": 4.6984461766570955e-06,
|
| 11922 |
+
"loss": 0.33419573307037354,
|
| 11923 |
+
"mean_token_accuracy": 0.8996593952178955,
|
| 11924 |
+
"num_tokens": 6643378.0,
|
| 11925 |
+
"step": 11420
|
| 11926 |
+
},
|
| 11927 |
+
{
|
| 11928 |
+
"entropy": 0.2496833168901503,
|
| 11929 |
+
"epoch": 0.8659254910130874,
|
| 11930 |
+
"grad_norm": 0.4080503582954407,
|
| 11931 |
+
"learning_rate": 4.646662991518775e-06,
|
| 11932 |
+
"loss": 0.26940312385559084,
|
| 11933 |
+
"mean_token_accuracy": 0.9210411250591278,
|
| 11934 |
+
"num_tokens": 6734073.0,
|
| 11935 |
+
"step": 11430
|
| 11936 |
+
},
|
| 11937 |
+
{
|
| 11938 |
+
"entropy": 0.3179155085235834,
|
| 11939 |
+
"epoch": 0.8666830811189606,
|
| 11940 |
+
"grad_norm": 0.35618963837623596,
|
| 11941 |
+
"learning_rate": 4.595152840828099e-06,
|
| 11942 |
+
"loss": 0.37588114738464357,
|
| 11943 |
+
"mean_token_accuracy": 0.8984893754124641,
|
| 11944 |
+
"num_tokens": 6843787.0,
|
| 11945 |
+
"step": 11440
|
| 11946 |
+
},
|
| 11947 |
+
{
|
| 11948 |
+
"entropy": 0.28430836256593467,
|
| 11949 |
+
"epoch": 0.8674406712248338,
|
| 11950 |
+
"grad_norm": 0.4345318675041199,
|
| 11951 |
+
"learning_rate": 4.543916034684531e-06,
|
| 11952 |
+
"loss": 0.33840193748474123,
|
| 11953 |
+
"mean_token_accuracy": 0.9102910846471787,
|
| 11954 |
+
"num_tokens": 6943533.0,
|
| 11955 |
+
"step": 11450
|
| 11956 |
+
},
|
| 11957 |
+
{
|
| 11958 |
+
"entropy": 0.28566361106932164,
|
| 11959 |
+
"epoch": 0.868198261330707,
|
| 11960 |
+
"grad_norm": 0.3477196991443634,
|
| 11961 |
+
"learning_rate": 4.492952881541962e-06,
|
| 11962 |
+
"loss": 0.34356863498687745,
|
| 11963 |
+
"mean_token_accuracy": 0.9089460060000419,
|
| 11964 |
+
"num_tokens": 7046514.0,
|
| 11965 |
+
"step": 11460
|
| 11966 |
+
},
|
| 11967 |
+
{
|
| 11968 |
+
"entropy": 0.2749630717560649,
|
| 11969 |
+
"epoch": 0.8689558514365803,
|
| 11970 |
+
"grad_norm": 0.44582805037498474,
|
| 11971 |
+
"learning_rate": 4.442263688206838e-06,
|
| 11972 |
+
"loss": 0.3119127988815308,
|
| 11973 |
+
"mean_token_accuracy": 0.9123926997184754,
|
| 11974 |
+
"num_tokens": 7147898.0,
|
| 11975 |
+
"step": 11470
|
| 11976 |
+
},
|
| 11977 |
+
{
|
| 11978 |
+
"entropy": 0.2725351193919778,
|
| 11979 |
+
"epoch": 0.8697134415424534,
|
| 11980 |
+
"grad_norm": 0.5893859267234802,
|
| 11981 |
+
"learning_rate": 4.391848759836331e-06,
|
| 11982 |
+
"loss": 0.30458645820617675,
|
| 11983 |
+
"mean_token_accuracy": 0.9120975598692894,
|
| 11984 |
+
"num_tokens": 7238758.0,
|
| 11985 |
+
"step": 11480
|
| 11986 |
+
},
|
| 11987 |
+
{
|
| 11988 |
+
"entropy": 0.2887442596256733,
|
| 11989 |
+
"epoch": 0.8704710316483266,
|
| 11990 |
+
"grad_norm": 0.38634830713272095,
|
| 11991 |
+
"learning_rate": 4.341708399936484e-06,
|
| 11992 |
+
"loss": 0.34334683418273926,
|
| 11993 |
+
"mean_token_accuracy": 0.9080741614103317,
|
| 11994 |
+
"num_tokens": 7332778.0,
|
| 11995 |
+
"step": 11490
|
| 11996 |
+
},
|
| 11997 |
+
{
|
| 11998 |
+
"entropy": 0.3163892447948456,
|
| 11999 |
+
"epoch": 0.8712286217541999,
|
| 12000 |
+
"grad_norm": 0.4110091030597687,
|
| 12001 |
+
"learning_rate": 4.291842910360411e-06,
|
| 12002 |
+
"loss": 0.3650186061859131,
|
| 12003 |
+
"mean_token_accuracy": 0.8998959764838219,
|
| 12004 |
+
"num_tokens": 7435633.0,
|
| 12005 |
+
"step": 11500
|
| 12006 |
+
},
|
| 12007 |
+
{
|
| 12008 |
+
"epoch": 0.8712286217541999,
|
| 12009 |
+
"eval_entropy": 0.28902850807940916,
|
| 12010 |
+
"eval_loss": 0.28520920872688293,
|
| 12011 |
+
"eval_mean_token_accuracy": 0.9075907013389501,
|
| 12012 |
+
"eval_num_tokens": 7435633.0,
|
| 12013 |
+
"eval_runtime": 362.4662,
|
| 12014 |
+
"eval_samples_per_second": 5.887,
|
| 12015 |
+
"eval_steps_per_second": 1.473,
|
| 12016 |
+
"step": 11500
|
| 12017 |
+
},
|
| 12018 |
+
{
|
| 12019 |
+
"entropy": 0.3062978575006127,
|
| 12020 |
+
"epoch": 0.8719862118600731,
|
| 12021 |
+
"grad_norm": 0.523257315158844,
|
| 12022 |
+
"learning_rate": 4.242252591306439e-06,
|
| 12023 |
+
"loss": 0.35051255226135253,
|
| 12024 |
+
"mean_token_accuracy": 0.9012857481837273,
|
| 12025 |
+
"num_tokens": 7532984.0,
|
| 12026 |
+
"step": 11510
|
| 12027 |
+
},
|
| 12028 |
+
{
|
| 12029 |
+
"entropy": 0.27845307141542436,
|
| 12030 |
+
"epoch": 0.8727438019659464,
|
| 12031 |
+
"grad_norm": 0.41053506731987,
|
| 12032 |
+
"learning_rate": 4.1929377413163386e-06,
|
| 12033 |
+
"loss": 0.33397209644317627,
|
| 12034 |
+
"mean_token_accuracy": 0.9132984727621078,
|
| 12035 |
+
"num_tokens": 7630708.0,
|
| 12036 |
+
"step": 11520
|
| 12037 |
+
},
|
| 12038 |
+
{
|
| 12039 |
+
"entropy": 0.2766347942873836,
|
| 12040 |
+
"epoch": 0.8735013920718195,
|
| 12041 |
+
"grad_norm": 0.3962479829788208,
|
| 12042 |
+
"learning_rate": 4.143898657273509e-06,
|
| 12043 |
+
"loss": 0.32157082557678224,
|
| 12044 |
+
"mean_token_accuracy": 0.9120013162493705,
|
| 12045 |
+
"num_tokens": 7731367.0,
|
| 12046 |
+
"step": 11530
|
| 12047 |
+
},
|
| 12048 |
+
{
|
| 12049 |
+
"entropy": 0.30036948043853046,
|
| 12050 |
+
"epoch": 0.8742589821776927,
|
| 12051 |
+
"grad_norm": 0.3602987825870514,
|
| 12052 |
+
"learning_rate": 4.095135634401198e-06,
|
| 12053 |
+
"loss": 0.3608391284942627,
|
| 12054 |
+
"mean_token_accuracy": 0.9042830646038056,
|
| 12055 |
+
"num_tokens": 7828677.0,
|
| 12056 |
+
"step": 11540
|
| 12057 |
+
},
|
| 12058 |
+
{
|
| 12059 |
+
"entropy": 0.28864908814430235,
|
| 12060 |
+
"epoch": 0.875016572283566,
|
| 12061 |
+
"grad_norm": 0.4127624034881592,
|
| 12062 |
+
"learning_rate": 4.046648966260708e-06,
|
| 12063 |
+
"loss": 0.33952603340148924,
|
| 12064 |
+
"mean_token_accuracy": 0.9077737912535667,
|
| 12065 |
+
"num_tokens": 7926097.0,
|
| 12066 |
+
"step": 11550
|
| 12067 |
+
},
|
| 12068 |
+
{
|
| 12069 |
+
"entropy": 0.28967538885772226,
|
| 12070 |
+
"epoch": 0.8757741623894392,
|
| 12071 |
+
"grad_norm": 0.3790387511253357,
|
| 12072 |
+
"learning_rate": 3.99843894474965e-06,
|
| 12073 |
+
"loss": 0.3468469619750977,
|
| 12074 |
+
"mean_token_accuracy": 0.9093071848154068,
|
| 12075 |
+
"num_tokens": 8024723.0,
|
| 12076 |
+
"step": 11560
|
| 12077 |
+
},
|
| 12078 |
+
{
|
| 12079 |
+
"entropy": 0.28250670302659275,
|
| 12080 |
+
"epoch": 0.8765317524953125,
|
| 12081 |
+
"grad_norm": 0.43996334075927734,
|
| 12082 |
+
"learning_rate": 3.950505860100173e-06,
|
| 12083 |
+
"loss": 0.31765334606170653,
|
| 12084 |
+
"mean_token_accuracy": 0.9129352971911431,
|
| 12085 |
+
"num_tokens": 8121274.0,
|
| 12086 |
+
"step": 11570
|
| 12087 |
+
},
|
| 12088 |
+
{
|
| 12089 |
+
"entropy": 0.2830861508846283,
|
| 12090 |
+
"epoch": 0.8772893426011856,
|
| 12091 |
+
"grad_norm": 0.49645763635635376,
|
| 12092 |
+
"learning_rate": 3.902850000877228e-06,
|
| 12093 |
+
"loss": 0.34697833061218264,
|
| 12094 |
+
"mean_token_accuracy": 0.9117490366101265,
|
| 12095 |
+
"num_tokens": 8214273.0,
|
| 12096 |
+
"step": 11580
|
| 12097 |
+
},
|
| 12098 |
+
{
|
| 12099 |
+
"entropy": 0.2739479450508952,
|
| 12100 |
+
"epoch": 0.8780469327070588,
|
| 12101 |
+
"grad_norm": 0.40956977009773254,
|
| 12102 |
+
"learning_rate": 3.855471653976833e-06,
|
| 12103 |
+
"loss": 0.31866281032562255,
|
| 12104 |
+
"mean_token_accuracy": 0.9111287161707878,
|
| 12105 |
+
"num_tokens": 8311017.0,
|
| 12106 |
+
"step": 11590
|
| 12107 |
+
},
|
| 12108 |
+
{
|
| 12109 |
+
"entropy": 0.24923466965556146,
|
| 12110 |
+
"epoch": 0.8788045228129321,
|
| 12111 |
+
"grad_norm": 0.46503928303718567,
|
| 12112 |
+
"learning_rate": 3.8083711046242976e-06,
|
| 12113 |
+
"loss": 0.29086575508117674,
|
| 12114 |
+
"mean_token_accuracy": 0.9209815084934234,
|
| 12115 |
+
"num_tokens": 8398430.0,
|
| 12116 |
+
"step": 11600
|
| 12117 |
+
},
|
| 12118 |
+
{
|
| 12119 |
+
"entropy": 0.2712684884667397,
|
| 12120 |
+
"epoch": 0.8795621129188053,
|
| 12121 |
+
"grad_norm": 0.5817049145698547,
|
| 12122 |
+
"learning_rate": 3.7615486363725862e-06,
|
| 12123 |
+
"loss": 0.317930793762207,
|
| 12124 |
+
"mean_token_accuracy": 0.9129366233944893,
|
| 12125 |
+
"num_tokens": 8499764.0,
|
| 12126 |
+
"step": 11610
|
| 12127 |
+
},
|
| 12128 |
+
{
|
| 12129 |
+
"entropy": 0.2838691871613264,
|
| 12130 |
+
"epoch": 0.8803197030246785,
|
| 12131 |
+
"grad_norm": 0.4317021667957306,
|
| 12132 |
+
"learning_rate": 3.715004531100552e-06,
|
| 12133 |
+
"loss": 0.33291192054748536,
|
| 12134 |
+
"mean_token_accuracy": 0.9116261765360832,
|
| 12135 |
+
"num_tokens": 8597456.0,
|
| 12136 |
+
"step": 11620
|
| 12137 |
+
},
|
| 12138 |
+
{
|
| 12139 |
+
"entropy": 0.3134252499789,
|
| 12140 |
+
"epoch": 0.8810772931305517,
|
| 12141 |
+
"grad_norm": 0.39608034491539,
|
| 12142 |
+
"learning_rate": 3.668739069011262e-06,
|
| 12143 |
+
"loss": 0.35954484939575193,
|
| 12144 |
+
"mean_token_accuracy": 0.9001775801181793,
|
| 12145 |
+
"num_tokens": 8695484.0,
|
| 12146 |
+
"step": 11630
|
| 12147 |
+
},
|
| 12148 |
+
{
|
| 12149 |
+
"entropy": 0.30583129059523345,
|
| 12150 |
+
"epoch": 0.8818348832364249,
|
| 12151 |
+
"grad_norm": 0.39231637120246887,
|
| 12152 |
+
"learning_rate": 3.6227525286303077e-06,
|
| 12153 |
+
"loss": 0.3442234516143799,
|
| 12154 |
+
"mean_token_accuracy": 0.9026809856295586,
|
| 12155 |
+
"num_tokens": 8797939.0,
|
| 12156 |
+
"step": 11640
|
| 12157 |
+
},
|
| 12158 |
+
{
|
| 12159 |
+
"entropy": 0.2677101884037256,
|
| 12160 |
+
"epoch": 0.8825924733422982,
|
| 12161 |
+
"grad_norm": 0.4724090099334717,
|
| 12162 |
+
"learning_rate": 3.5770451868041176e-06,
|
| 12163 |
+
"loss": 0.3116743087768555,
|
| 12164 |
+
"mean_token_accuracy": 0.9159694716334343,
|
| 12165 |
+
"num_tokens": 8888698.0,
|
| 12166 |
+
"step": 11650
|
| 12167 |
+
},
|
| 12168 |
+
{
|
| 12169 |
+
"entropy": 0.2846052870154381,
|
| 12170 |
+
"epoch": 0.8833500634481714,
|
| 12171 |
+
"grad_norm": 0.4535260796546936,
|
| 12172 |
+
"learning_rate": 3.5316173186983047e-06,
|
| 12173 |
+
"loss": 0.3254629850387573,
|
| 12174 |
+
"mean_token_accuracy": 0.9086816072463989,
|
| 12175 |
+
"num_tokens": 8984200.0,
|
| 12176 |
+
"step": 11660
|
| 12177 |
+
},
|
| 12178 |
+
{
|
| 12179 |
+
"entropy": 0.2747473681345582,
|
| 12180 |
+
"epoch": 0.8841076535540446,
|
| 12181 |
+
"grad_norm": 0.4237743616104126,
|
| 12182 |
+
"learning_rate": 3.486469197796016e-06,
|
| 12183 |
+
"loss": 0.3034130811691284,
|
| 12184 |
+
"mean_token_accuracy": 0.91242136657238,
|
| 12185 |
+
"num_tokens": 9081499.0,
|
| 12186 |
+
"step": 11670
|
| 12187 |
+
},
|
| 12188 |
+
{
|
| 12189 |
+
"entropy": 0.29838810246437786,
|
| 12190 |
+
"epoch": 0.8848652436599178,
|
| 12191 |
+
"grad_norm": 0.46438512206077576,
|
| 12192 |
+
"learning_rate": 3.4416010958962575e-06,
|
| 12193 |
+
"loss": 0.3530242443084717,
|
| 12194 |
+
"mean_token_accuracy": 0.903732942044735,
|
| 12195 |
+
"num_tokens": 9180324.0,
|
| 12196 |
+
"step": 11680
|
| 12197 |
+
},
|
| 12198 |
+
{
|
| 12199 |
+
"entropy": 0.2972015455365181,
|
| 12200 |
+
"epoch": 0.885622833765791,
|
| 12201 |
+
"grad_norm": 0.5626263618469238,
|
| 12202 |
+
"learning_rate": 3.3970132831122825e-06,
|
| 12203 |
+
"loss": 0.3368989944458008,
|
| 12204 |
+
"mean_token_accuracy": 0.9042138263583184,
|
| 12205 |
+
"num_tokens": 9281689.0,
|
| 12206 |
+
"step": 11690
|
| 12207 |
+
},
|
| 12208 |
+
{
|
| 12209 |
+
"entropy": 0.28716162145137786,
|
| 12210 |
+
"epoch": 0.8863804238716643,
|
| 12211 |
+
"grad_norm": 0.4129055440425873,
|
| 12212 |
+
"learning_rate": 3.3527060278699563e-06,
|
| 12213 |
+
"loss": 0.333894681930542,
|
| 12214 |
+
"mean_token_accuracy": 0.9115474045276641,
|
| 12215 |
+
"num_tokens": 9373442.0,
|
| 12216 |
+
"step": 11700
|
| 12217 |
+
},
|
| 12218 |
+
{
|
| 12219 |
+
"entropy": 0.3078194234520197,
|
| 12220 |
+
"epoch": 0.8871380139775374,
|
| 12221 |
+
"grad_norm": 0.3562985956668854,
|
| 12222 |
+
"learning_rate": 3.3086795969061557e-06,
|
| 12223 |
+
"loss": 0.3490109443664551,
|
| 12224 |
+
"mean_token_accuracy": 0.9009092181921006,
|
| 12225 |
+
"num_tokens": 9474012.0,
|
| 12226 |
+
"step": 11710
|
| 12227 |
+
},
|
| 12228 |
+
{
|
| 12229 |
+
"entropy": 0.32325697001069786,
|
| 12230 |
+
"epoch": 0.8878956040834106,
|
| 12231 |
+
"grad_norm": 0.4092463552951813,
|
| 12232 |
+
"learning_rate": 3.264934255267127e-06,
|
| 12233 |
+
"loss": 0.38159396648406985,
|
| 12234 |
+
"mean_token_accuracy": 0.894842478632927,
|
| 12235 |
+
"num_tokens": 9584832.0,
|
| 12236 |
+
"step": 11720
|
| 12237 |
+
},
|
| 12238 |
+
{
|
| 12239 |
+
"entropy": 0.24773544836789368,
|
| 12240 |
+
"epoch": 0.8886531941892839,
|
| 12241 |
+
"grad_norm": 0.43235835433006287,
|
| 12242 |
+
"learning_rate": 3.2214702663069308e-06,
|
| 12243 |
+
"loss": 0.2853692054748535,
|
| 12244 |
+
"mean_token_accuracy": 0.9219878166913986,
|
| 12245 |
+
"num_tokens": 9685029.0,
|
| 12246 |
+
"step": 11730
|
| 12247 |
+
},
|
| 12248 |
+
{
|
| 12249 |
+
"entropy": 0.2398093307390809,
|
| 12250 |
+
"epoch": 0.8894107842951571,
|
| 12251 |
+
"grad_norm": 0.35412466526031494,
|
| 12252 |
+
"learning_rate": 3.1782878916858273e-06,
|
| 12253 |
+
"loss": 0.29156792163848877,
|
| 12254 |
+
"mean_token_accuracy": 0.9237233802676201,
|
| 12255 |
+
"num_tokens": 9781772.0,
|
| 12256 |
+
"step": 11740
|
| 12257 |
+
},
|
| 12258 |
+
{
|
| 12259 |
+
"entropy": 0.2823848981410265,
|
| 12260 |
+
"epoch": 0.8901683744010304,
|
| 12261 |
+
"grad_norm": 0.4498985707759857,
|
| 12262 |
+
"learning_rate": 3.1353873913687327e-06,
|
| 12263 |
+
"loss": 0.3437998294830322,
|
| 12264 |
+
"mean_token_accuracy": 0.907787038385868,
|
| 12265 |
+
"num_tokens": 9874032.0,
|
| 12266 |
+
"step": 11750
|
| 12267 |
+
},
|
| 12268 |
+
{
|
| 12269 |
+
"epoch": 0.8901683744010304,
|
| 12270 |
+
"eval_entropy": 0.2877783386964588,
|
| 12271 |
+
"eval_loss": 0.28504320979118347,
|
| 12272 |
+
"eval_mean_token_accuracy": 0.907661068528779,
|
| 12273 |
+
"eval_num_tokens": 9874032.0,
|
| 12274 |
+
"eval_runtime": 364.3306,
|
| 12275 |
+
"eval_samples_per_second": 5.857,
|
| 12276 |
+
"eval_steps_per_second": 1.466,
|
| 12277 |
+
"step": 11750
|
| 12278 |
+
},
|
| 12279 |
+
{
|
| 12280 |
+
"entropy": 0.26369482185691595,
|
| 12281 |
+
"epoch": 0.8909259645069035,
|
| 12282 |
+
"grad_norm": 0.5122780203819275,
|
| 12283 |
+
"learning_rate": 3.092769023623604e-06,
|
| 12284 |
+
"loss": 0.3001707077026367,
|
| 12285 |
+
"mean_token_accuracy": 0.9151472270488739,
|
| 12286 |
+
"num_tokens": 9969565.0,
|
| 12287 |
+
"step": 11760
|
| 12288 |
+
},
|
| 12289 |
+
{
|
| 12290 |
+
"entropy": 0.2945669298991561,
|
| 12291 |
+
"epoch": 0.8916835546127767,
|
| 12292 |
+
"grad_norm": 0.44441863894462585,
|
| 12293 |
+
"learning_rate": 3.050433045019935e-06,
|
| 12294 |
+
"loss": 0.34853763580322267,
|
| 12295 |
+
"mean_token_accuracy": 0.9060156434774399,
|
| 12296 |
+
"num_tokens": 10073446.0,
|
| 12297 |
+
"step": 11770
|
| 12298 |
+
},
|
| 12299 |
+
{
|
| 12300 |
+
"entropy": 0.28827795311808585,
|
| 12301 |
+
"epoch": 0.89244114471865,
|
| 12302 |
+
"grad_norm": 0.44912177324295044,
|
| 12303 |
+
"learning_rate": 3.008379710427184e-06,
|
| 12304 |
+
"loss": 0.3416128635406494,
|
| 12305 |
+
"mean_token_accuracy": 0.9073840662837028,
|
| 12306 |
+
"num_tokens": 10178033.0,
|
| 12307 |
+
"step": 11780
|
| 12308 |
+
},
|
| 12309 |
+
{
|
| 12310 |
+
"entropy": 0.29140637572854755,
|
| 12311 |
+
"epoch": 0.8931987348245232,
|
| 12312 |
+
"grad_norm": 0.42318063974380493,
|
| 12313 |
+
"learning_rate": 2.9666092730132546e-06,
|
| 12314 |
+
"loss": 0.32835893630981444,
|
| 12315 |
+
"mean_token_accuracy": 0.9083481669425965,
|
| 12316 |
+
"num_tokens": 10279691.0,
|
| 12317 |
+
"step": 11790
|
| 12318 |
+
},
|
| 12319 |
+
{
|
| 12320 |
+
"entropy": 0.2990898171439767,
|
| 12321 |
+
"epoch": 0.8939563249303965,
|
| 12322 |
+
"grad_norm": 0.5038789510726929,
|
| 12323 |
+
"learning_rate": 2.925121984242951e-06,
|
| 12324 |
+
"loss": 0.34421396255493164,
|
| 12325 |
+
"mean_token_accuracy": 0.9057884082198143,
|
| 12326 |
+
"num_tokens": 10375462.0,
|
| 12327 |
+
"step": 11800
|
| 12328 |
+
},
|
| 12329 |
+
{
|
| 12330 |
+
"entropy": 0.26596365328878163,
|
| 12331 |
+
"epoch": 0.8947139150362696,
|
| 12332 |
+
"grad_norm": 0.5523534417152405,
|
| 12333 |
+
"learning_rate": 2.8839180938764863e-06,
|
| 12334 |
+
"loss": 0.31238632202148436,
|
| 12335 |
+
"mean_token_accuracy": 0.9166907787322998,
|
| 12336 |
+
"num_tokens": 10468250.0,
|
| 12337 |
+
"step": 11810
|
| 12338 |
+
},
|
| 12339 |
+
{
|
| 12340 |
+
"entropy": 0.26575090661644934,
|
| 12341 |
+
"epoch": 0.8954715051421428,
|
| 12342 |
+
"grad_norm": 0.4703275263309479,
|
| 12343 |
+
"learning_rate": 2.842997849967971e-06,
|
| 12344 |
+
"loss": 0.29630722999572756,
|
| 12345 |
+
"mean_token_accuracy": 0.9156646102666854,
|
| 12346 |
+
"num_tokens": 10563173.0,
|
| 12347 |
+
"step": 11820
|
| 12348 |
+
},
|
| 12349 |
+
{
|
| 12350 |
+
"entropy": 0.2932529145851731,
|
| 12351 |
+
"epoch": 0.8962290952480161,
|
| 12352 |
+
"grad_norm": 0.3833015263080597,
|
| 12353 |
+
"learning_rate": 2.8023614988639037e-06,
|
| 12354 |
+
"loss": 0.3358645439147949,
|
| 12355 |
+
"mean_token_accuracy": 0.9068357229232789,
|
| 12356 |
+
"num_tokens": 10664872.0,
|
| 12357 |
+
"step": 11830
|
| 12358 |
+
},
|
| 12359 |
+
{
|
| 12360 |
+
"entropy": 0.3053524810820818,
|
| 12361 |
+
"epoch": 0.8969866853538893,
|
| 12362 |
+
"grad_norm": 0.48305702209472656,
|
| 12363 |
+
"learning_rate": 2.7620092852017163e-06,
|
| 12364 |
+
"loss": 0.3639130830764771,
|
| 12365 |
+
"mean_token_accuracy": 0.8995088264346123,
|
| 12366 |
+
"num_tokens": 10765468.0,
|
| 12367 |
+
"step": 11840
|
| 12368 |
+
},
|
| 12369 |
+
{
|
| 12370 |
+
"entropy": 0.27784552425146103,
|
| 12371 |
+
"epoch": 0.8977442754597625,
|
| 12372 |
+
"grad_norm": 0.7603539228439331,
|
| 12373 |
+
"learning_rate": 2.721941451908283e-06,
|
| 12374 |
+
"loss": 0.3374777317047119,
|
| 12375 |
+
"mean_token_accuracy": 0.9113835170865059,
|
| 12376 |
+
"num_tokens": 10860042.0,
|
| 12377 |
+
"step": 11850
|
| 12378 |
+
},
|
| 12379 |
+
{
|
| 12380 |
+
"entropy": 0.2647241912782192,
|
| 12381 |
+
"epoch": 0.8985018655656357,
|
| 12382 |
+
"grad_norm": 0.41474252939224243,
|
| 12383 |
+
"learning_rate": 2.6821582401984624e-06,
|
| 12384 |
+
"loss": 0.3280783653259277,
|
| 12385 |
+
"mean_token_accuracy": 0.9178823977708817,
|
| 12386 |
+
"num_tokens": 10956561.0,
|
| 12387 |
+
"step": 11860
|
| 12388 |
+
},
|
| 12389 |
+
{
|
| 12390 |
+
"entropy": 0.29947108291089536,
|
| 12391 |
+
"epoch": 0.8992594556715089,
|
| 12392 |
+
"grad_norm": 0.7285243272781372,
|
| 12393 |
+
"learning_rate": 2.6426598895736397e-06,
|
| 12394 |
+
"loss": 0.3578646659851074,
|
| 12395 |
+
"mean_token_accuracy": 0.9039269968867302,
|
| 12396 |
+
"num_tokens": 11052288.0,
|
| 12397 |
+
"step": 11870
|
| 12398 |
+
},
|
| 12399 |
+
{
|
| 12400 |
+
"entropy": 0.2672554237768054,
|
| 12401 |
+
"epoch": 0.9000170457773822,
|
| 12402 |
+
"grad_norm": 0.3953443765640259,
|
| 12403 |
+
"learning_rate": 2.6034466378203006e-06,
|
| 12404 |
+
"loss": 0.32422146797180174,
|
| 12405 |
+
"mean_token_accuracy": 0.9144565418362618,
|
| 12406 |
+
"num_tokens": 11148839.0,
|
| 12407 |
+
"step": 11880
|
| 12408 |
+
},
|
| 12409 |
+
{
|
| 12410 |
+
"entropy": 0.31262458153069017,
|
| 12411 |
+
"epoch": 0.9007746358832553,
|
| 12412 |
+
"grad_norm": 0.41961008310317993,
|
| 12413 |
+
"learning_rate": 2.56451872100858e-06,
|
| 12414 |
+
"loss": 0.36020898818969727,
|
| 12415 |
+
"mean_token_accuracy": 0.8983094453811645,
|
| 12416 |
+
"num_tokens": 11249056.0,
|
| 12417 |
+
"step": 11890
|
| 12418 |
+
},
|
| 12419 |
+
{
|
| 12420 |
+
"entropy": 0.2922690697014332,
|
| 12421 |
+
"epoch": 0.9015322259891286,
|
| 12422 |
+
"grad_norm": 0.435486763715744,
|
| 12423 |
+
"learning_rate": 2.5258763734908543e-06,
|
| 12424 |
+
"loss": 0.33933568000793457,
|
| 12425 |
+
"mean_token_accuracy": 0.9072558209300041,
|
| 12426 |
+
"num_tokens": 11350465.0,
|
| 12427 |
+
"step": 11900
|
| 12428 |
+
},
|
| 12429 |
+
{
|
| 12430 |
+
"entropy": 0.2990702610462904,
|
| 12431 |
+
"epoch": 0.9022898160950018,
|
| 12432 |
+
"grad_norm": 0.3435786962509155,
|
| 12433 |
+
"learning_rate": 2.4875198279003197e-06,
|
| 12434 |
+
"loss": 0.32688493728637696,
|
| 12435 |
+
"mean_token_accuracy": 0.9039709404110908,
|
| 12436 |
+
"num_tokens": 11461689.0,
|
| 12437 |
+
"step": 11910
|
| 12438 |
+
},
|
| 12439 |
+
{
|
| 12440 |
+
"entropy": 0.2897825580090284,
|
| 12441 |
+
"epoch": 0.903047406200875,
|
| 12442 |
+
"grad_norm": 0.43031200766563416,
|
| 12443 |
+
"learning_rate": 2.4494493151496102e-06,
|
| 12444 |
+
"loss": 0.3513793468475342,
|
| 12445 |
+
"mean_token_accuracy": 0.9052729144692421,
|
| 12446 |
+
"num_tokens": 11552766.0,
|
| 12447 |
+
"step": 11920
|
| 12448 |
+
},
|
| 12449 |
+
{
|
| 12450 |
+
"entropy": 0.3014989258721471,
|
| 12451 |
+
"epoch": 0.9038049963067483,
|
| 12452 |
+
"grad_norm": 0.5037825703620911,
|
| 12453 |
+
"learning_rate": 2.4116650644293915e-06,
|
| 12454 |
+
"loss": 0.34794044494628906,
|
| 12455 |
+
"mean_token_accuracy": 0.9035784974694252,
|
| 12456 |
+
"num_tokens": 11649054.0,
|
| 12457 |
+
"step": 11930
|
| 12458 |
+
},
|
| 12459 |
+
{
|
| 12460 |
+
"entropy": 0.26335182003676894,
|
| 12461 |
+
"epoch": 0.9045625864126214,
|
| 12462 |
+
"grad_norm": 0.4968317449092865,
|
| 12463 |
+
"learning_rate": 2.3741673032069756e-06,
|
| 12464 |
+
"loss": 0.29862327575683595,
|
| 12465 |
+
"mean_token_accuracy": 0.916699244081974,
|
| 12466 |
+
"num_tokens": 11745904.0,
|
| 12467 |
+
"step": 11940
|
| 12468 |
+
},
|
| 12469 |
+
{
|
| 12470 |
+
"entropy": 0.27952721528708935,
|
| 12471 |
+
"epoch": 0.9053201765184947,
|
| 12472 |
+
"grad_norm": 0.4142918586730957,
|
| 12473 |
+
"learning_rate": 2.336956257224976e-06,
|
| 12474 |
+
"loss": 0.3438222646713257,
|
| 12475 |
+
"mean_token_accuracy": 0.9101695835590362,
|
| 12476 |
+
"num_tokens": 11848622.0,
|
| 12477 |
+
"step": 11950
|
| 12478 |
+
},
|
| 12479 |
+
{
|
| 12480 |
+
"entropy": 0.2757990902289748,
|
| 12481 |
+
"epoch": 0.9060777666243679,
|
| 12482 |
+
"grad_norm": 0.6232902407646179,
|
| 12483 |
+
"learning_rate": 2.3000321504999256e-06,
|
| 12484 |
+
"loss": 0.301646089553833,
|
| 12485 |
+
"mean_token_accuracy": 0.9144108861684799,
|
| 12486 |
+
"num_tokens": 11948966.0,
|
| 12487 |
+
"step": 11960
|
| 12488 |
+
},
|
| 12489 |
+
{
|
| 12490 |
+
"entropy": 0.29781850054860115,
|
| 12491 |
+
"epoch": 0.9068353567302411,
|
| 12492 |
+
"grad_norm": 0.5357040166854858,
|
| 12493 |
+
"learning_rate": 2.2633952053209396e-06,
|
| 12494 |
+
"loss": 0.3362586498260498,
|
| 12495 |
+
"mean_token_accuracy": 0.9043088600039482,
|
| 12496 |
+
"num_tokens": 12050900.0,
|
| 12497 |
+
"step": 11970
|
| 12498 |
+
},
|
| 12499 |
+
{
|
| 12500 |
+
"entropy": 0.3077027928084135,
|
| 12501 |
+
"epoch": 0.9075929468361144,
|
| 12502 |
+
"grad_norm": 0.3803006708621979,
|
| 12503 |
+
"learning_rate": 2.2270456422483655e-06,
|
| 12504 |
+
"loss": 0.36307554244995116,
|
| 12505 |
+
"mean_token_accuracy": 0.8973830282688141,
|
| 12506 |
+
"num_tokens": 12147023.0,
|
| 12507 |
+
"step": 11980
|
| 12508 |
+
},
|
| 12509 |
+
{
|
| 12510 |
+
"entropy": 0.2707913937047124,
|
| 12511 |
+
"epoch": 0.9083505369419875,
|
| 12512 |
+
"grad_norm": 0.47196221351623535,
|
| 12513 |
+
"learning_rate": 2.190983680112485e-06,
|
| 12514 |
+
"loss": 0.31296596527099607,
|
| 12515 |
+
"mean_token_accuracy": 0.9124700129032135,
|
| 12516 |
+
"num_tokens": 12235151.0,
|
| 12517 |
+
"step": 11990
|
| 12518 |
+
},
|
| 12519 |
+
{
|
| 12520 |
+
"entropy": 0.2599682783707976,
|
| 12521 |
+
"epoch": 0.9091081270478607,
|
| 12522 |
+
"grad_norm": 0.49874573945999146,
|
| 12523 |
+
"learning_rate": 2.155209536012148e-06,
|
| 12524 |
+
"loss": 0.33502559661865233,
|
| 12525 |
+
"mean_token_accuracy": 0.9186246544122696,
|
| 12526 |
+
"num_tokens": 12323993.0,
|
| 12527 |
+
"step": 12000
|
| 12528 |
+
},
|
| 12529 |
+
{
|
| 12530 |
+
"epoch": 0.9091081270478607,
|
| 12531 |
+
"eval_entropy": 0.2891411053553949,
|
| 12532 |
+
"eval_loss": 0.28493180871009827,
|
| 12533 |
+
"eval_mean_token_accuracy": 0.907693214184336,
|
| 12534 |
+
"eval_num_tokens": 12323993.0,
|
| 12535 |
+
"eval_runtime": 362.6666,
|
| 12536 |
+
"eval_samples_per_second": 5.884,
|
| 12537 |
+
"eval_steps_per_second": 1.472,
|
| 12538 |
+
"step": 12000
|
| 12539 |
+
},
|
| 12540 |
+
{
|
| 12541 |
+
"entropy": 0.3315456384792924,
|
| 12542 |
+
"epoch": 0.909865717153734,
|
| 12543 |
+
"grad_norm": 0.36325132846832275,
|
| 12544 |
+
"learning_rate": 2.119723425313508e-06,
|
| 12545 |
+
"loss": 0.3699875593185425,
|
| 12546 |
+
"mean_token_accuracy": 0.8940782964229583,
|
| 12547 |
+
"num_tokens": 12430861.0,
|
| 12548 |
+
"step": 12010
|
| 12549 |
+
},
|
| 12550 |
+
{
|
| 12551 |
+
"entropy": 0.3365255404263735,
|
| 12552 |
+
"epoch": 0.9106233072596072,
|
| 12553 |
+
"grad_norm": 0.4198094308376312,
|
| 12554 |
+
"learning_rate": 2.0845255616487146e-06,
|
| 12555 |
+
"loss": 0.39371466636657715,
|
| 12556 |
+
"mean_token_accuracy": 0.8894189238548279,
|
| 12557 |
+
"num_tokens": 12538044.0,
|
| 12558 |
+
"step": 12020
|
| 12559 |
+
},
|
| 12560 |
+
{
|
| 12561 |
+
"entropy": 0.28438792023807763,
|
| 12562 |
+
"epoch": 0.9113808973654804,
|
| 12563 |
+
"grad_norm": 0.39958682656288147,
|
| 12564 |
+
"learning_rate": 2.049616156914619e-06,
|
| 12565 |
+
"loss": 0.32168402671813967,
|
| 12566 |
+
"mean_token_accuracy": 0.9102310806512832,
|
| 12567 |
+
"num_tokens": 12639720.0,
|
| 12568 |
+
"step": 12030
|
| 12569 |
+
},
|
| 12570 |
+
{
|
| 12571 |
+
"entropy": 0.2975167462602258,
|
| 12572 |
+
"epoch": 0.9121384874713536,
|
| 12573 |
+
"grad_norm": 0.43647003173828125,
|
| 12574 |
+
"learning_rate": 2.014995421271498e-06,
|
| 12575 |
+
"loss": 0.34835057258605956,
|
| 12576 |
+
"mean_token_accuracy": 0.9067213013768196,
|
| 12577 |
+
"num_tokens": 12743172.0,
|
| 12578 |
+
"step": 12040
|
| 12579 |
+
},
|
| 12580 |
+
{
|
| 12581 |
+
"entropy": 0.28764260224997995,
|
| 12582 |
+
"epoch": 0.9128960775772268,
|
| 12583 |
+
"grad_norm": 0.3330994248390198,
|
| 12584 |
+
"learning_rate": 1.9806635631418048e-06,
|
| 12585 |
+
"loss": 0.3269298315048218,
|
| 12586 |
+
"mean_token_accuracy": 0.9144092231988907,
|
| 12587 |
+
"num_tokens": 12843459.0,
|
| 12588 |
+
"step": 12050
|
| 12589 |
+
},
|
| 12590 |
+
{
|
| 12591 |
+
"entropy": 0.28370366655290125,
|
| 12592 |
+
"epoch": 0.9136536676831001,
|
| 12593 |
+
"grad_norm": 0.46237102150917053,
|
| 12594 |
+
"learning_rate": 1.9466207892088805e-06,
|
| 12595 |
+
"loss": 0.32587761878967286,
|
| 12596 |
+
"mean_token_accuracy": 0.9086109101772308,
|
| 12597 |
+
"num_tokens": 12949924.0,
|
| 12598 |
+
"step": 12060
|
| 12599 |
+
},
|
| 12600 |
+
{
|
| 12601 |
+
"entropy": 0.31762309623882173,
|
| 12602 |
+
"epoch": 0.9144112577889733,
|
| 12603 |
+
"grad_norm": 0.4568561315536499,
|
| 12604 |
+
"learning_rate": 1.912867304415761e-06,
|
| 12605 |
+
"loss": 0.37861361503601076,
|
| 12606 |
+
"mean_token_accuracy": 0.898035791516304,
|
| 12607 |
+
"num_tokens": 13054862.0,
|
| 12608 |
+
"step": 12070
|
| 12609 |
+
},
|
| 12610 |
+
{
|
| 12611 |
+
"entropy": 0.29679893162101506,
|
| 12612 |
+
"epoch": 0.9151688478948465,
|
| 12613 |
+
"grad_norm": 0.4890940487384796,
|
| 12614 |
+
"learning_rate": 1.8794033119638953e-06,
|
| 12615 |
+
"loss": 0.3435173988342285,
|
| 12616 |
+
"mean_token_accuracy": 0.9060111805796623,
|
| 12617 |
+
"num_tokens": 13159877.0,
|
| 12618 |
+
"step": 12080
|
| 12619 |
+
},
|
| 12620 |
+
{
|
| 12621 |
+
"entropy": 0.26280158162117007,
|
| 12622 |
+
"epoch": 0.9159264380007197,
|
| 12623 |
+
"grad_norm": 0.48253440856933594,
|
| 12624 |
+
"learning_rate": 1.846229013311951e-06,
|
| 12625 |
+
"loss": 0.29681944847106934,
|
| 12626 |
+
"mean_token_accuracy": 0.9169836983084678,
|
| 12627 |
+
"num_tokens": 13250560.0,
|
| 12628 |
+
"step": 12090
|
| 12629 |
+
},
|
| 12630 |
+
{
|
| 12631 |
+
"entropy": 0.3261014189571142,
|
| 12632 |
+
"epoch": 0.9166840281065929,
|
| 12633 |
+
"grad_norm": 0.4755929708480835,
|
| 12634 |
+
"learning_rate": 1.8133446081745886e-06,
|
| 12635 |
+
"loss": 0.36304349899291993,
|
| 12636 |
+
"mean_token_accuracy": 0.8956771790981293,
|
| 12637 |
+
"num_tokens": 13352805.0,
|
| 12638 |
+
"step": 12100
|
| 12639 |
+
},
|
| 12640 |
+
{
|
| 12641 |
+
"entropy": 0.27531050983816385,
|
| 12642 |
+
"epoch": 0.9174416182124662,
|
| 12643 |
+
"grad_norm": 0.4745216965675354,
|
| 12644 |
+
"learning_rate": 1.7807502945212605e-06,
|
| 12645 |
+
"loss": 0.3325106859207153,
|
| 12646 |
+
"mean_token_accuracy": 0.9112949371337891,
|
| 12647 |
+
"num_tokens": 13447502.0,
|
| 12648 |
+
"step": 12110
|
| 12649 |
+
},
|
| 12650 |
+
{
|
| 12651 |
+
"entropy": 0.26114649400115014,
|
| 12652 |
+
"epoch": 0.9181992083183393,
|
| 12653 |
+
"grad_norm": 0.4469934403896332,
|
| 12654 |
+
"learning_rate": 1.748446268575027e-06,
|
| 12655 |
+
"loss": 0.30356202125549314,
|
| 12656 |
+
"mean_token_accuracy": 0.916273832321167,
|
| 12657 |
+
"num_tokens": 13542441.0,
|
| 12658 |
+
"step": 12120
|
| 12659 |
+
},
|
| 12660 |
+
{
|
| 12661 |
+
"entropy": 0.2954039676114917,
|
| 12662 |
+
"epoch": 0.9189567984242126,
|
| 12663 |
+
"grad_norm": 0.4029696583747864,
|
| 12664 |
+
"learning_rate": 1.7164327248113622e-06,
|
| 12665 |
+
"loss": 0.33674476146697996,
|
| 12666 |
+
"mean_token_accuracy": 0.906862536072731,
|
| 12667 |
+
"num_tokens": 13650072.0,
|
| 12668 |
+
"step": 12130
|
| 12669 |
+
},
|
| 12670 |
+
{
|
| 12671 |
+
"entropy": 0.24167130887508392,
|
| 12672 |
+
"epoch": 0.9197143885300858,
|
| 12673 |
+
"grad_norm": 0.44493207335472107,
|
| 12674 |
+
"learning_rate": 1.6847098559569952e-06,
|
| 12675 |
+
"loss": 0.26566371917724607,
|
| 12676 |
+
"mean_token_accuracy": 0.923855085670948,
|
| 12677 |
+
"num_tokens": 13744399.0,
|
| 12678 |
+
"step": 12140
|
| 12679 |
+
},
|
| 12680 |
+
{
|
| 12681 |
+
"entropy": 0.2633110383525491,
|
| 12682 |
+
"epoch": 0.920471978635959,
|
| 12683 |
+
"grad_norm": 0.451709121465683,
|
| 12684 |
+
"learning_rate": 1.653277852988744e-06,
|
| 12685 |
+
"loss": 0.313076114654541,
|
| 12686 |
+
"mean_token_accuracy": 0.9192310333251953,
|
| 12687 |
+
"num_tokens": 13837719.0,
|
| 12688 |
+
"step": 12150
|
| 12689 |
+
},
|
| 12690 |
+
{
|
| 12691 |
+
"entropy": 0.29435538705438374,
|
| 12692 |
+
"epoch": 0.9212295687418323,
|
| 12693 |
+
"grad_norm": 0.39204367995262146,
|
| 12694 |
+
"learning_rate": 1.6221369051323655e-06,
|
| 12695 |
+
"loss": 0.3549853563308716,
|
| 12696 |
+
"mean_token_accuracy": 0.9049896076321602,
|
| 12697 |
+
"num_tokens": 13931355.0,
|
| 12698 |
+
"step": 12160
|
| 12699 |
+
},
|
| 12700 |
+
{
|
| 12701 |
+
"entropy": 0.29209454338997604,
|
| 12702 |
+
"epoch": 0.9219871588477054,
|
| 12703 |
+
"grad_norm": 0.38774025440216064,
|
| 12704 |
+
"learning_rate": 1.5912871998614187e-06,
|
| 12705 |
+
"loss": 0.34168984889984133,
|
| 12706 |
+
"mean_token_accuracy": 0.9070258140563965,
|
| 12707 |
+
"num_tokens": 14039359.0,
|
| 12708 |
+
"step": 12170
|
| 12709 |
+
},
|
| 12710 |
+
{
|
| 12711 |
+
"entropy": 0.31331349443644285,
|
| 12712 |
+
"epoch": 0.9227447489535787,
|
| 12713 |
+
"grad_norm": 0.41252318024635315,
|
| 12714 |
+
"learning_rate": 1.5607289228961374e-06,
|
| 12715 |
+
"loss": 0.3593226432800293,
|
| 12716 |
+
"mean_token_accuracy": 0.8983490958809852,
|
| 12717 |
+
"num_tokens": 14137741.0,
|
| 12718 |
+
"step": 12180
|
| 12719 |
+
},
|
| 12720 |
+
{
|
| 12721 |
+
"entropy": 0.28924416061490776,
|
| 12722 |
+
"epoch": 0.9235023390594519,
|
| 12723 |
+
"grad_norm": 0.5740947127342224,
|
| 12724 |
+
"learning_rate": 1.5304622582023031e-06,
|
| 12725 |
+
"loss": 0.3327749252319336,
|
| 12726 |
+
"mean_token_accuracy": 0.9081077992916107,
|
| 12727 |
+
"num_tokens": 14232465.0,
|
| 12728 |
+
"step": 12190
|
| 12729 |
+
},
|
| 12730 |
+
{
|
| 12731 |
+
"entropy": 0.2850802129134536,
|
| 12732 |
+
"epoch": 0.9242599291653251,
|
| 12733 |
+
"grad_norm": 0.4056800305843353,
|
| 12734 |
+
"learning_rate": 1.5004873879901571e-06,
|
| 12735 |
+
"loss": 0.34341509342193605,
|
| 12736 |
+
"mean_token_accuracy": 0.9104896888136864,
|
| 12737 |
+
"num_tokens": 14326738.0,
|
| 12738 |
+
"step": 12200
|
| 12739 |
+
},
|
| 12740 |
+
{
|
| 12741 |
+
"entropy": 0.2863912934437394,
|
| 12742 |
+
"epoch": 0.9250175192711984,
|
| 12743 |
+
"grad_norm": 0.3346513509750366,
|
| 12744 |
+
"learning_rate": 1.4708044927132737e-06,
|
| 12745 |
+
"loss": 0.35355463027954104,
|
| 12746 |
+
"mean_token_accuracy": 0.9096838116645813,
|
| 12747 |
+
"num_tokens": 14424392.0,
|
| 12748 |
+
"step": 12210
|
| 12749 |
+
},
|
| 12750 |
+
{
|
| 12751 |
+
"entropy": 0.3078114558942616,
|
| 12752 |
+
"epoch": 0.9257751093770715,
|
| 12753 |
+
"grad_norm": 0.38695788383483887,
|
| 12754 |
+
"learning_rate": 1.441413751067494e-06,
|
| 12755 |
+
"loss": 0.367154598236084,
|
| 12756 |
+
"mean_token_accuracy": 0.9017126500606537,
|
| 12757 |
+
"num_tokens": 14534354.0,
|
| 12758 |
+
"step": 12220
|
| 12759 |
+
},
|
| 12760 |
+
{
|
| 12761 |
+
"entropy": 0.32155216187238694,
|
| 12762 |
+
"epoch": 0.9265326994829447,
|
| 12763 |
+
"grad_norm": 0.5330353379249573,
|
| 12764 |
+
"learning_rate": 1.4123153399898614e-06,
|
| 12765 |
+
"loss": 0.36757822036743165,
|
| 12766 |
+
"mean_token_accuracy": 0.8965023845434189,
|
| 12767 |
+
"num_tokens": 14639036.0,
|
| 12768 |
+
"step": 12230
|
| 12769 |
+
},
|
| 12770 |
+
{
|
| 12771 |
+
"entropy": 0.2916434150189161,
|
| 12772 |
+
"epoch": 0.927290289588818,
|
| 12773 |
+
"grad_norm": 0.46392399072647095,
|
| 12774 |
+
"learning_rate": 1.3835094346575262e-06,
|
| 12775 |
+
"loss": 0.33872079849243164,
|
| 12776 |
+
"mean_token_accuracy": 0.9077073231339454,
|
| 12777 |
+
"num_tokens": 14740021.0,
|
| 12778 |
+
"step": 12240
|
| 12779 |
+
},
|
| 12780 |
+
{
|
| 12781 |
+
"entropy": 0.2472118590027094,
|
| 12782 |
+
"epoch": 0.9280478796946912,
|
| 12783 |
+
"grad_norm": 0.37260034680366516,
|
| 12784 |
+
"learning_rate": 1.3549962084867197e-06,
|
| 12785 |
+
"loss": 0.29043848514556886,
|
| 12786 |
+
"mean_token_accuracy": 0.9208429232239723,
|
| 12787 |
+
"num_tokens": 14830662.0,
|
| 12788 |
+
"step": 12250
|
| 12789 |
+
},
|
| 12790 |
+
{
|
| 12791 |
+
"epoch": 0.9280478796946912,
|
| 12792 |
+
"eval_entropy": 0.2879207830890064,
|
| 12793 |
+
"eval_loss": 0.2848566472530365,
|
| 12794 |
+
"eval_mean_token_accuracy": 0.9077223223470123,
|
| 12795 |
+
"eval_num_tokens": 14830662.0,
|
| 12796 |
+
"eval_runtime": 362.6256,
|
| 12797 |
+
"eval_samples_per_second": 5.885,
|
| 12798 |
+
"eval_steps_per_second": 1.473,
|
| 12799 |
+
"step": 12250
|
| 12800 |
+
},
|
| 12801 |
+
{
|
| 12802 |
+
"entropy": 0.27810837496072055,
|
| 12803 |
+
"epoch": 0.9288054698005644,
|
| 12804 |
+
"grad_norm": 0.4152062237262726,
|
| 12805 |
+
"learning_rate": 1.326775833131677e-06,
|
| 12806 |
+
"loss": 0.34612598419189455,
|
| 12807 |
+
"mean_token_accuracy": 0.9133247256278991,
|
| 12808 |
+
"num_tokens": 14925605.0,
|
| 12809 |
+
"step": 12260
|
| 12810 |
+
},
|
| 12811 |
+
{
|
| 12812 |
+
"entropy": 0.2930800588801503,
|
| 12813 |
+
"epoch": 0.9295630599064376,
|
| 12814 |
+
"grad_norm": 0.3420519530773163,
|
| 12815 |
+
"learning_rate": 1.2988484784836497e-06,
|
| 12816 |
+
"loss": 0.3541224956512451,
|
| 12817 |
+
"mean_token_accuracy": 0.9066302254796028,
|
| 12818 |
+
"num_tokens": 15025948.0,
|
| 12819 |
+
"step": 12270
|
| 12820 |
+
},
|
| 12821 |
+
{
|
| 12822 |
+
"entropy": 0.34085101950913665,
|
| 12823 |
+
"epoch": 0.9303206500123108,
|
| 12824 |
+
"grad_norm": 0.42782142758369446,
|
| 12825 |
+
"learning_rate": 1.2712143126698384e-06,
|
| 12826 |
+
"loss": 0.40104336738586427,
|
| 12827 |
+
"mean_token_accuracy": 0.8878615260124206,
|
| 12828 |
+
"num_tokens": 15137971.0,
|
| 12829 |
+
"step": 12280
|
| 12830 |
+
},
|
| 12831 |
+
{
|
| 12832 |
+
"entropy": 0.24184838123619556,
|
| 12833 |
+
"epoch": 0.9310782401181841,
|
| 12834 |
+
"grad_norm": 0.4608571231365204,
|
| 12835 |
+
"learning_rate": 1.2438735020524062e-06,
|
| 12836 |
+
"loss": 0.2901769161224365,
|
| 12837 |
+
"mean_token_accuracy": 0.9232783406972885,
|
| 12838 |
+
"num_tokens": 15234313.0,
|
| 12839 |
+
"step": 12290
|
| 12840 |
+
},
|
| 12841 |
+
{
|
| 12842 |
+
"entropy": 0.2717073464766145,
|
| 12843 |
+
"epoch": 0.9318358302240572,
|
| 12844 |
+
"grad_norm": 0.46019455790519714,
|
| 12845 |
+
"learning_rate": 1.2168262112274676e-06,
|
| 12846 |
+
"loss": 0.3081354141235352,
|
| 12847 |
+
"mean_token_accuracy": 0.9139297619462013,
|
| 12848 |
+
"num_tokens": 15331902.0,
|
| 12849 |
+
"step": 12300
|
| 12850 |
+
},
|
| 12851 |
+
{
|
| 12852 |
+
"entropy": 0.3171485211700201,
|
| 12853 |
+
"epoch": 0.9325934203299305,
|
| 12854 |
+
"grad_norm": 0.44192010164260864,
|
| 12855 |
+
"learning_rate": 1.1900726030241005e-06,
|
| 12856 |
+
"loss": 0.359798789024353,
|
| 12857 |
+
"mean_token_accuracy": 0.899483498930931,
|
| 12858 |
+
"num_tokens": 15431160.0,
|
| 12859 |
+
"step": 12310
|
| 12860 |
+
},
|
| 12861 |
+
{
|
| 12862 |
+
"entropy": 0.29771914295852187,
|
| 12863 |
+
"epoch": 0.9333510104358037,
|
| 12864 |
+
"grad_norm": 0.35575592517852783,
|
| 12865 |
+
"learning_rate": 1.1636128385033696e-06,
|
| 12866 |
+
"loss": 0.32083702087402344,
|
| 12867 |
+
"mean_token_accuracy": 0.9065295130014419,
|
| 12868 |
+
"num_tokens": 15528945.0,
|
| 12869 |
+
"step": 12320
|
| 12870 |
+
},
|
| 12871 |
+
{
|
| 12872 |
+
"entropy": 0.328072508983314,
|
| 12873 |
+
"epoch": 0.9341086005416769,
|
| 12874 |
+
"grad_norm": 0.37298858165740967,
|
| 12875 |
+
"learning_rate": 1.1374470769573541e-06,
|
| 12876 |
+
"loss": 0.3967781066894531,
|
| 12877 |
+
"mean_token_accuracy": 0.8941646784543991,
|
| 12878 |
+
"num_tokens": 15630573.0,
|
| 12879 |
+
"step": 12330
|
| 12880 |
+
},
|
| 12881 |
+
{
|
| 12882 |
+
"entropy": 0.30953355710953473,
|
| 12883 |
+
"epoch": 0.9348661906475502,
|
| 12884 |
+
"grad_norm": 0.4739098846912384,
|
| 12885 |
+
"learning_rate": 1.1115754759081777e-06,
|
| 12886 |
+
"loss": 0.3533921480178833,
|
| 12887 |
+
"mean_token_accuracy": 0.8996481195092201,
|
| 12888 |
+
"num_tokens": 15734620.0,
|
| 12889 |
+
"step": 12340
|
| 12890 |
+
},
|
| 12891 |
+
{
|
| 12892 |
+
"entropy": 0.30768233593553307,
|
| 12893 |
+
"epoch": 0.9356237807534233,
|
| 12894 |
+
"grad_norm": 0.4143330156803131,
|
| 12895 |
+
"learning_rate": 1.0859981911070793e-06,
|
| 12896 |
+
"loss": 0.33932952880859374,
|
| 12897 |
+
"mean_token_accuracy": 0.9027550190687179,
|
| 12898 |
+
"num_tokens": 15833310.0,
|
| 12899 |
+
"step": 12350
|
| 12900 |
+
},
|
| 12901 |
+
{
|
| 12902 |
+
"entropy": 0.2843058746308088,
|
| 12903 |
+
"epoch": 0.9363813708592966,
|
| 12904 |
+
"grad_norm": 0.4079853296279907,
|
| 12905 |
+
"learning_rate": 1.060715376533472e-06,
|
| 12906 |
+
"loss": 0.3476103782653809,
|
| 12907 |
+
"mean_token_accuracy": 0.9069626480340958,
|
| 12908 |
+
"num_tokens": 15933036.0,
|
| 12909 |
+
"step": 12360
|
| 12910 |
+
},
|
| 12911 |
+
{
|
| 12912 |
+
"entropy": 0.2918760960921645,
|
| 12913 |
+
"epoch": 0.9371389609651698,
|
| 12914 |
+
"grad_norm": 0.3709755539894104,
|
| 12915 |
+
"learning_rate": 1.0357271843939864e-06,
|
| 12916 |
+
"loss": 0.32259211540222166,
|
| 12917 |
+
"mean_token_accuracy": 0.9056548669934272,
|
| 12918 |
+
"num_tokens": 16035404.0,
|
| 12919 |
+
"step": 12370
|
| 12920 |
+
},
|
| 12921 |
+
{
|
| 12922 |
+
"entropy": 0.2681634811684489,
|
| 12923 |
+
"epoch": 0.937896551071043,
|
| 12924 |
+
"grad_norm": 0.5038941502571106,
|
| 12925 |
+
"learning_rate": 1.0110337651215995e-06,
|
| 12926 |
+
"loss": 0.31412277221679685,
|
| 12927 |
+
"mean_token_accuracy": 0.915467070043087,
|
| 12928 |
+
"num_tokens": 16132094.0,
|
| 12929 |
+
"step": 12380
|
| 12930 |
+
},
|
| 12931 |
+
{
|
| 12932 |
+
"entropy": 0.311520766839385,
|
| 12933 |
+
"epoch": 0.9386541411769163,
|
| 12934 |
+
"grad_norm": 0.42714932560920715,
|
| 12935 |
+
"learning_rate": 9.866352673747025e-07,
|
| 12936 |
+
"loss": 0.3627468109130859,
|
| 12937 |
+
"mean_token_accuracy": 0.9007305800914764,
|
| 12938 |
+
"num_tokens": 16224173.0,
|
| 12939 |
+
"step": 12390
|
| 12940 |
+
},
|
| 12941 |
+
{
|
| 12942 |
+
"entropy": 0.24273829255253077,
|
| 12943 |
+
"epoch": 0.9394117312827894,
|
| 12944 |
+
"grad_norm": 0.4695494472980499,
|
| 12945 |
+
"learning_rate": 9.62531838036218e-07,
|
| 12946 |
+
"loss": 0.2682880163192749,
|
| 12947 |
+
"mean_token_accuracy": 0.9242408692836761,
|
| 12948 |
+
"num_tokens": 16318335.0,
|
| 12949 |
+
"step": 12400
|
| 12950 |
+
},
|
| 12951 |
+
{
|
| 12952 |
+
"entropy": 0.2871857823804021,
|
| 12953 |
+
"epoch": 0.9401693213886627,
|
| 12954 |
+
"grad_norm": 0.3833423852920532,
|
| 12955 |
+
"learning_rate": 9.387236222126894e-07,
|
| 12956 |
+
"loss": 0.32558517456054686,
|
| 12957 |
+
"mean_token_accuracy": 0.9094377249479294,
|
| 12958 |
+
"num_tokens": 16417299.0,
|
| 12959 |
+
"step": 12410
|
| 12960 |
+
},
|
| 12961 |
+
{
|
| 12962 |
+
"entropy": 0.3126089680939913,
|
| 12963 |
+
"epoch": 0.9409269114945359,
|
| 12964 |
+
"grad_norm": 0.41594237089157104,
|
| 12965 |
+
"learning_rate": 9.152107632334428e-07,
|
| 12966 |
+
"loss": 0.3656469345092773,
|
| 12967 |
+
"mean_token_accuracy": 0.8993785634636879,
|
| 12968 |
+
"num_tokens": 16516774.0,
|
| 12969 |
+
"step": 12420
|
| 12970 |
+
},
|
| 12971 |
+
{
|
| 12972 |
+
"entropy": 0.2751727135851979,
|
| 12973 |
+
"epoch": 0.9416845016004091,
|
| 12974 |
+
"grad_norm": 0.5532023310661316,
|
| 12975 |
+
"learning_rate": 8.9199340264971e-07,
|
| 12976 |
+
"loss": 0.3194823503494263,
|
| 12977 |
+
"mean_token_accuracy": 0.9125665321946144,
|
| 12978 |
+
"num_tokens": 16618168.0,
|
| 12979 |
+
"step": 12430
|
| 12980 |
+
},
|
| 12981 |
+
{
|
| 12982 |
+
"entropy": 0.2662426060065627,
|
| 12983 |
+
"epoch": 0.9424420917062823,
|
| 12984 |
+
"grad_norm": 0.42721033096313477,
|
| 12985 |
+
"learning_rate": 8.690716802337628e-07,
|
| 12986 |
+
"loss": 0.32259302139282225,
|
| 12987 |
+
"mean_token_accuracy": 0.9164244115352631,
|
| 12988 |
+
"num_tokens": 16717553.0,
|
| 12989 |
+
"step": 12440
|
| 12990 |
+
},
|
| 12991 |
+
{
|
| 12992 |
+
"entropy": 0.2952677818015218,
|
| 12993 |
+
"epoch": 0.9431996818121555,
|
| 12994 |
+
"grad_norm": 0.5415988564491272,
|
| 12995 |
+
"learning_rate": 8.464457339780907e-07,
|
| 12996 |
+
"loss": 0.35180134773254396,
|
| 12997 |
+
"mean_token_accuracy": 0.9077468425035476,
|
| 12998 |
+
"num_tokens": 16803125.0,
|
| 12999 |
+
"step": 12450
|
| 13000 |
+
},
|
| 13001 |
+
{
|
| 13002 |
+
"entropy": 0.27926060985773804,
|
| 13003 |
+
"epoch": 0.9439572719180287,
|
| 13004 |
+
"grad_norm": 0.5080968141555786,
|
| 13005 |
+
"learning_rate": 8.241157000945631e-07,
|
| 13006 |
+
"loss": 0.31758029460906984,
|
| 13007 |
+
"mean_token_accuracy": 0.9103856220841408,
|
| 13008 |
+
"num_tokens": 16902308.0,
|
| 13009 |
+
"step": 12460
|
| 13010 |
+
},
|
| 13011 |
+
{
|
| 13012 |
+
"entropy": 0.29949437715113164,
|
| 13013 |
+
"epoch": 0.944714862023902,
|
| 13014 |
+
"grad_norm": 0.5077760815620422,
|
| 13015 |
+
"learning_rate": 8.020817130136083e-07,
|
| 13016 |
+
"loss": 0.3514946460723877,
|
| 13017 |
+
"mean_token_accuracy": 0.9059733942151069,
|
| 13018 |
+
"num_tokens": 17003776.0,
|
| 13019 |
+
"step": 12470
|
| 13020 |
+
},
|
| 13021 |
+
{
|
| 13022 |
+
"entropy": 0.30533884670585393,
|
| 13023 |
+
"epoch": 0.9454724521297752,
|
| 13024 |
+
"grad_norm": 0.47848737239837646,
|
| 13025 |
+
"learning_rate": 7.803439053833961e-07,
|
| 13026 |
+
"loss": 0.36388406753540037,
|
| 13027 |
+
"mean_token_accuracy": 0.9001722037792206,
|
| 13028 |
+
"num_tokens": 17104375.0,
|
| 13029 |
+
"step": 12480
|
| 13030 |
+
},
|
| 13031 |
+
{
|
| 13032 |
+
"entropy": 0.32481123935431244,
|
| 13033 |
+
"epoch": 0.9462300422356484,
|
| 13034 |
+
"grad_norm": 0.3776695728302002,
|
| 13035 |
+
"learning_rate": 7.589024080690621e-07,
|
| 13036 |
+
"loss": 0.3672971725463867,
|
| 13037 |
+
"mean_token_accuracy": 0.8965393304824829,
|
| 13038 |
+
"num_tokens": 17210665.0,
|
| 13039 |
+
"step": 12490
|
| 13040 |
+
},
|
| 13041 |
+
{
|
| 13042 |
+
"entropy": 0.3082854451611638,
|
| 13043 |
+
"epoch": 0.9469876323415216,
|
| 13044 |
+
"grad_norm": 0.3830999732017517,
|
| 13045 |
+
"learning_rate": 7.377573501519019e-07,
|
| 13046 |
+
"loss": 0.34308252334594724,
|
| 13047 |
+
"mean_token_accuracy": 0.9049276232719421,
|
| 13048 |
+
"num_tokens": 17308912.0,
|
| 13049 |
+
"step": 12500
|
| 13050 |
+
},
|
| 13051 |
+
{
|
| 13052 |
+
"epoch": 0.9469876323415216,
|
| 13053 |
+
"eval_entropy": 0.28848998702030054,
|
| 13054 |
+
"eval_loss": 0.2848212718963623,
|
| 13055 |
+
"eval_mean_token_accuracy": 0.9077069452416138,
|
| 13056 |
+
"eval_num_tokens": 17308912.0,
|
| 13057 |
+
"eval_runtime": 363.3483,
|
| 13058 |
+
"eval_samples_per_second": 5.873,
|
| 13059 |
+
"eval_steps_per_second": 1.47,
|
| 13060 |
+
"step": 12500
|
| 13061 |
}
|
| 13062 |
],
|
| 13063 |
"logging_steps": 10,
|
|
|
|
| 13077 |
"attributes": {}
|
| 13078 |
}
|
| 13079 |
},
|
| 13080 |
+
"total_flos": 1.0843380675373957e+18,
|
| 13081 |
"train_batch_size": 2,
|
| 13082 |
"trial_name": null,
|
| 13083 |
"trial_params": null
|