Training in progress, epoch 3, checkpoint
Browse files- last-checkpoint/model-00001-of-00003.safetensors +1 -1
- last-checkpoint/model-00002-of-00003.safetensors +1 -1
- last-checkpoint/model-00003-of-00003.safetensors +1 -1
- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/rng_state_4.pth +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +684 -4
last-checkpoint/model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4949453792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4cb217517103aa1f9b67671200e1088c82e998f977765551951ecbd6cacb9fc3
|
| 3 |
size 4949453792
|
last-checkpoint/model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999819336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d306a9a8a0ee55b4dfde88bec95cd9d2efca13601a22479321364d39624178e8
|
| 3 |
size 4999819336
|
last-checkpoint/model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4546807800
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9842fedf4985d9dd12c414781f60bc29e4a8029b6b63daa86ef5332b34e099e7
|
| 3 |
size 4546807800
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 28992348490
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09fbba3766d6cb31a87875e81304d345ce1b8197cb449721c0f15cd8e1db6251
|
| 3 |
size 28992348490
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb18ac8d6db3307b1c242f7cb069fc8b8dab957434ddfcafcac997cfd6a43abf
|
| 3 |
size 15984
|
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 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4bdab708057b5f34a402d9a2b4443f5f93a8e8ee2ddb66d955f0a15ad394ecc5
|
| 3 |
size 15984
|
last-checkpoint/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:599882a30c163a5a2a000c4e74b320ecc4a55aa1b079882fd66aa3d2559d19e7
|
| 3 |
size 15984
|
last-checkpoint/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:567c3b482c209c2778fc017e39a38642c488edda20673ef29f571ef7177ad81e
|
| 3 |
size 15984
|
last-checkpoint/rng_state_4.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f9ffe9a916e778423aaed4ec842923c9ccfdd3d7a4fbad10dc6a3bfc278fb8e
|
| 3 |
size 15984
|
last-checkpoint/rng_state_5.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7ede8a81aa3c780fb9c3cb57537752a782c4aed1dcecb7aafd6ca5a7ea90252
|
| 3 |
size 15984
|
last-checkpoint/rng_state_6.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b1c5c0c0afa907d332467e631e6cee80ba476689aa0caa77689ca273d83b3e4
|
| 3 |
size 15984
|
last-checkpoint/rng_state_7.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73025ac422abb13303ee974109cf39f6f848de7f7013e828d04aa4e2ec0e6757
|
| 3 |
size 15984
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4dbc6caf624eba5c924d315b74568bd01ca1fccdd670b0ff8efdc24821d15151
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 300,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1382,6 +1382,686 @@
|
|
| 1382 |
"learning_rate": 5.050432566228552e-07,
|
| 1383 |
"loss": 1.1831,
|
| 1384 |
"step": 960
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1385 |
}
|
| 1386 |
],
|
| 1387 |
"logging_steps": 5,
|
|
@@ -1396,12 +2076,12 @@
|
|
| 1396 |
"should_evaluate": false,
|
| 1397 |
"should_log": false,
|
| 1398 |
"should_save": true,
|
| 1399 |
-
"should_training_stop":
|
| 1400 |
},
|
| 1401 |
"attributes": {}
|
| 1402 |
}
|
| 1403 |
},
|
| 1404 |
-
"total_flos":
|
| 1405 |
"train_batch_size": 8,
|
| 1406 |
"trial_name": null,
|
| 1407 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 300,
|
| 6 |
+
"global_step": 1443,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1382 |
"learning_rate": 5.050432566228552e-07,
|
| 1383 |
"loss": 1.1831,
|
| 1384 |
"step": 960
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 2.006237006237006,
|
| 1388 |
+
"grad_norm": 3.609375,
|
| 1389 |
+
"learning_rate": 4.956010076471065e-07,
|
| 1390 |
+
"loss": 1.1592,
|
| 1391 |
+
"step": 965
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 2.0166320166320166,
|
| 1395 |
+
"grad_norm": 3.53125,
|
| 1396 |
+
"learning_rate": 4.862186938359441e-07,
|
| 1397 |
+
"loss": 1.1667,
|
| 1398 |
+
"step": 970
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 2.027027027027027,
|
| 1402 |
+
"grad_norm": 3.5625,
|
| 1403 |
+
"learning_rate": 4.768974300419573e-07,
|
| 1404 |
+
"loss": 1.1678,
|
| 1405 |
+
"step": 975
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 2.0374220374220373,
|
| 1409 |
+
"grad_norm": 3.625,
|
| 1410 |
+
"learning_rate": 4.6763832386347214e-07,
|
| 1411 |
+
"loss": 1.1723,
|
| 1412 |
+
"step": 980
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.047817047817048,
|
| 1416 |
+
"grad_norm": 3.578125,
|
| 1417 |
+
"learning_rate": 4.5844247551294224e-07,
|
| 1418 |
+
"loss": 1.1642,
|
| 1419 |
+
"step": 985
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.0582120582120584,
|
| 1423 |
+
"grad_norm": 3.609375,
|
| 1424 |
+
"learning_rate": 4.493109776862143e-07,
|
| 1425 |
+
"loss": 1.17,
|
| 1426 |
+
"step": 990
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 2.0686070686070686,
|
| 1430 |
+
"grad_norm": 3.640625,
|
| 1431 |
+
"learning_rate": 4.402449154326913e-07,
|
| 1432 |
+
"loss": 1.1651,
|
| 1433 |
+
"step": 995
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 2.079002079002079,
|
| 1437 |
+
"grad_norm": 3.640625,
|
| 1438 |
+
"learning_rate": 4.312453660263987e-07,
|
| 1439 |
+
"loss": 1.1783,
|
| 1440 |
+
"step": 1000
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 2.0893970893970892,
|
| 1444 |
+
"grad_norm": 3.578125,
|
| 1445 |
+
"learning_rate": 4.2231339883798025e-07,
|
| 1446 |
+
"loss": 1.1711,
|
| 1447 |
+
"step": 1005
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"epoch": 2.0997920997921,
|
| 1451 |
+
"grad_norm": 3.5625,
|
| 1452 |
+
"learning_rate": 4.13450075207628e-07,
|
| 1453 |
+
"loss": 1.163,
|
| 1454 |
+
"step": 1010
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.1101871101871104,
|
| 1458 |
+
"grad_norm": 3.515625,
|
| 1459 |
+
"learning_rate": 4.0465644831897006e-07,
|
| 1460 |
+
"loss": 1.17,
|
| 1461 |
+
"step": 1015
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 2.1205821205821205,
|
| 1465 |
+
"grad_norm": 3.5625,
|
| 1466 |
+
"learning_rate": 3.9593356307392436e-07,
|
| 1467 |
+
"loss": 1.1733,
|
| 1468 |
+
"step": 1020
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 2.130977130977131,
|
| 1472 |
+
"grad_norm": 3.59375,
|
| 1473 |
+
"learning_rate": 3.872824559685409e-07,
|
| 1474 |
+
"loss": 1.1762,
|
| 1475 |
+
"step": 1025
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 2.141372141372141,
|
| 1479 |
+
"grad_norm": 3.671875,
|
| 1480 |
+
"learning_rate": 3.7870415496983743e-07,
|
| 1481 |
+
"loss": 1.1734,
|
| 1482 |
+
"step": 1030
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 2.1517671517671517,
|
| 1486 |
+
"grad_norm": 3.578125,
|
| 1487 |
+
"learning_rate": 3.701996793936535e-07,
|
| 1488 |
+
"loss": 1.1724,
|
| 1489 |
+
"step": 1035
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 2.1621621621621623,
|
| 1493 |
+
"grad_norm": 3.5625,
|
| 1494 |
+
"learning_rate": 3.6177003978352917e-07,
|
| 1495 |
+
"loss": 1.1718,
|
| 1496 |
+
"step": 1040
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.1725571725571724,
|
| 1500 |
+
"grad_norm": 3.5625,
|
| 1501 |
+
"learning_rate": 3.5341623779062813e-07,
|
| 1502 |
+
"loss": 1.1688,
|
| 1503 |
+
"step": 1045
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 2.182952182952183,
|
| 1507 |
+
"grad_norm": 3.609375,
|
| 1508 |
+
"learning_rate": 3.45139266054715e-07,
|
| 1509 |
+
"loss": 1.1732,
|
| 1510 |
+
"step": 1050
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 2.1933471933471935,
|
| 1514 |
+
"grad_norm": 3.59375,
|
| 1515 |
+
"learning_rate": 3.3694010808620733e-07,
|
| 1516 |
+
"loss": 1.1619,
|
| 1517 |
+
"step": 1055
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 2.2037422037422036,
|
| 1521 |
+
"grad_norm": 3.625,
|
| 1522 |
+
"learning_rate": 3.288197381493075e-07,
|
| 1523 |
+
"loss": 1.1673,
|
| 1524 |
+
"step": 1060
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 2.214137214137214,
|
| 1528 |
+
"grad_norm": 3.609375,
|
| 1529 |
+
"learning_rate": 3.207791211462383e-07,
|
| 1530 |
+
"loss": 1.1725,
|
| 1531 |
+
"step": 1065
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 2.2245322245322248,
|
| 1535 |
+
"grad_norm": 3.546875,
|
| 1536 |
+
"learning_rate": 3.128192125025869e-07,
|
| 1537 |
+
"loss": 1.1673,
|
| 1538 |
+
"step": 1070
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.234927234927235,
|
| 1542 |
+
"grad_norm": 3.578125,
|
| 1543 |
+
"learning_rate": 3.049409580537773e-07,
|
| 1544 |
+
"loss": 1.1735,
|
| 1545 |
+
"step": 1075
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 2.2453222453222454,
|
| 1549 |
+
"grad_norm": 3.5625,
|
| 1550 |
+
"learning_rate": 2.9714529393268016e-07,
|
| 1551 |
+
"loss": 1.1583,
|
| 1552 |
+
"step": 1080
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 2.2557172557172556,
|
| 1556 |
+
"grad_norm": 3.484375,
|
| 1557 |
+
"learning_rate": 2.8943314645837955e-07,
|
| 1558 |
+
"loss": 1.1715,
|
| 1559 |
+
"step": 1085
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 2.266112266112266,
|
| 1563 |
+
"grad_norm": 3.609375,
|
| 1564 |
+
"learning_rate": 2.8180543202609984e-07,
|
| 1565 |
+
"loss": 1.164,
|
| 1566 |
+
"step": 1090
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 2.2765072765072767,
|
| 1570 |
+
"grad_norm": 3.53125,
|
| 1571 |
+
"learning_rate": 2.742630569983182e-07,
|
| 1572 |
+
"loss": 1.1695,
|
| 1573 |
+
"step": 1095
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 2.286902286902287,
|
| 1577 |
+
"grad_norm": 3.578125,
|
| 1578 |
+
"learning_rate": 2.66806917597064e-07,
|
| 1579 |
+
"loss": 1.169,
|
| 1580 |
+
"step": 1100
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.2972972972972974,
|
| 1584 |
+
"grad_norm": 3.578125,
|
| 1585 |
+
"learning_rate": 2.594378997974267e-07,
|
| 1586 |
+
"loss": 1.1615,
|
| 1587 |
+
"step": 1105
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 2.3076923076923075,
|
| 1591 |
+
"grad_norm": 3.6875,
|
| 1592 |
+
"learning_rate": 2.5215687922227845e-07,
|
| 1593 |
+
"loss": 1.1712,
|
| 1594 |
+
"step": 1110
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 2.318087318087318,
|
| 1598 |
+
"grad_norm": 3.515625,
|
| 1599 |
+
"learning_rate": 2.4496472103823027e-07,
|
| 1600 |
+
"loss": 1.1688,
|
| 1601 |
+
"step": 1115
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 2.3284823284823286,
|
| 1605 |
+
"grad_norm": 3.609375,
|
| 1606 |
+
"learning_rate": 2.378622798528266e-07,
|
| 1607 |
+
"loss": 1.1631,
|
| 1608 |
+
"step": 1120
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 2.3388773388773387,
|
| 1612 |
+
"grad_norm": 3.6875,
|
| 1613 |
+
"learning_rate": 2.3085039961299814e-07,
|
| 1614 |
+
"loss": 1.1671,
|
| 1615 |
+
"step": 1125
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 2.3492723492723493,
|
| 1619 |
+
"grad_norm": 3.640625,
|
| 1620 |
+
"learning_rate": 2.239299135047794e-07,
|
| 1621 |
+
"loss": 1.1623,
|
| 1622 |
+
"step": 1130
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 2.35966735966736,
|
| 1626 |
+
"grad_norm": 3.578125,
|
| 1627 |
+
"learning_rate": 2.1710164385430585e-07,
|
| 1628 |
+
"loss": 1.1716,
|
| 1629 |
+
"step": 1135
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 2.37006237006237,
|
| 1633 |
+
"grad_norm": 3.546875,
|
| 1634 |
+
"learning_rate": 2.103664020300997e-07,
|
| 1635 |
+
"loss": 1.1674,
|
| 1636 |
+
"step": 1140
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 2.3804573804573805,
|
| 1640 |
+
"grad_norm": 3.453125,
|
| 1641 |
+
"learning_rate": 2.037249883466614e-07,
|
| 1642 |
+
"loss": 1.1623,
|
| 1643 |
+
"step": 1145
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 2.390852390852391,
|
| 1647 |
+
"grad_norm": 3.65625,
|
| 1648 |
+
"learning_rate": 1.971781919693697e-07,
|
| 1649 |
+
"loss": 1.1808,
|
| 1650 |
+
"step": 1150
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 2.401247401247401,
|
| 1654 |
+
"grad_norm": 3.625,
|
| 1655 |
+
"learning_rate": 1.9072679082071163e-07,
|
| 1656 |
+
"loss": 1.169,
|
| 1657 |
+
"step": 1155
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 2.4116424116424118,
|
| 1661 |
+
"grad_norm": 3.671875,
|
| 1662 |
+
"learning_rate": 1.8437155148784433e-07,
|
| 1663 |
+
"loss": 1.1717,
|
| 1664 |
+
"step": 1160
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.422037422037422,
|
| 1668 |
+
"grad_norm": 3.59375,
|
| 1669 |
+
"learning_rate": 1.781132291315064e-07,
|
| 1670 |
+
"loss": 1.1706,
|
| 1671 |
+
"step": 1165
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 2.4324324324324325,
|
| 1675 |
+
"grad_norm": 3.625,
|
| 1676 |
+
"learning_rate": 1.7195256739628439e-07,
|
| 1677 |
+
"loss": 1.1722,
|
| 1678 |
+
"step": 1170
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 2.442827442827443,
|
| 1682 |
+
"grad_norm": 3.71875,
|
| 1683 |
+
"learning_rate": 1.6589029832225155e-07,
|
| 1684 |
+
"loss": 1.1615,
|
| 1685 |
+
"step": 1175
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 2.453222453222453,
|
| 1689 |
+
"grad_norm": 3.625,
|
| 1690 |
+
"learning_rate": 1.599271422579812e-07,
|
| 1691 |
+
"loss": 1.1691,
|
| 1692 |
+
"step": 1180
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 2.4636174636174637,
|
| 1696 |
+
"grad_norm": 3.640625,
|
| 1697 |
+
"learning_rate": 1.5406380777495297e-07,
|
| 1698 |
+
"loss": 1.1647,
|
| 1699 |
+
"step": 1185
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 2.474012474012474,
|
| 1703 |
+
"grad_norm": 3.59375,
|
| 1704 |
+
"learning_rate": 1.4830099158335563e-07,
|
| 1705 |
+
"loss": 1.1707,
|
| 1706 |
+
"step": 1190
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 2.4844074844074844,
|
| 1710 |
+
"grad_norm": 3.578125,
|
| 1711 |
+
"learning_rate": 1.426393784493015e-07,
|
| 1712 |
+
"loss": 1.1564,
|
| 1713 |
+
"step": 1195
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 2.494802494802495,
|
| 1717 |
+
"grad_norm": 3.59375,
|
| 1718 |
+
"learning_rate": 1.3707964111345805e-07,
|
| 1719 |
+
"loss": 1.1721,
|
| 1720 |
+
"step": 1200
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 2.494802494802495,
|
| 1724 |
+
"eval_loss": 1.1930803060531616,
|
| 1725 |
+
"eval_runtime": 10.8355,
|
| 1726 |
+
"eval_samples_per_second": 85.829,
|
| 1727 |
+
"eval_steps_per_second": 2.769,
|
| 1728 |
+
"step": 1200
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 2.505197505197505,
|
| 1732 |
+
"grad_norm": 3.65625,
|
| 1733 |
+
"learning_rate": 1.3162244021111123e-07,
|
| 1734 |
+
"loss": 1.1677,
|
| 1735 |
+
"step": 1205
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 2.5155925155925156,
|
| 1739 |
+
"grad_norm": 3.59375,
|
| 1740 |
+
"learning_rate": 1.2626842419366369e-07,
|
| 1741 |
+
"loss": 1.1551,
|
| 1742 |
+
"step": 1210
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 2.525987525987526,
|
| 1746 |
+
"grad_norm": 3.640625,
|
| 1747 |
+
"learning_rate": 1.2101822925158378e-07,
|
| 1748 |
+
"loss": 1.1678,
|
| 1749 |
+
"step": 1215
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 2.5363825363825363,
|
| 1753 |
+
"grad_norm": 3.65625,
|
| 1754 |
+
"learning_rate": 1.1587247923881016e-07,
|
| 1755 |
+
"loss": 1.1731,
|
| 1756 |
+
"step": 1220
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 2.546777546777547,
|
| 1760 |
+
"grad_norm": 3.6875,
|
| 1761 |
+
"learning_rate": 1.1083178559862227e-07,
|
| 1762 |
+
"loss": 1.1707,
|
| 1763 |
+
"step": 1225
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 2.5571725571725574,
|
| 1767 |
+
"grad_norm": 3.671875,
|
| 1768 |
+
"learning_rate": 1.0589674729098507e-07,
|
| 1769 |
+
"loss": 1.1733,
|
| 1770 |
+
"step": 1230
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 2.5675675675675675,
|
| 1774 |
+
"grad_norm": 3.578125,
|
| 1775 |
+
"learning_rate": 1.0106795072137896e-07,
|
| 1776 |
+
"loss": 1.1741,
|
| 1777 |
+
"step": 1235
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 2.577962577962578,
|
| 1781 |
+
"grad_norm": 3.625,
|
| 1782 |
+
"learning_rate": 9.634596967111853e-08,
|
| 1783 |
+
"loss": 1.1704,
|
| 1784 |
+
"step": 1240
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 2.5883575883575882,
|
| 1788 |
+
"grad_norm": 3.640625,
|
| 1789 |
+
"learning_rate": 9.173136522917457e-08,
|
| 1790 |
+
"loss": 1.1679,
|
| 1791 |
+
"step": 1245
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 2.598752598752599,
|
| 1795 |
+
"grad_norm": 3.546875,
|
| 1796 |
+
"learning_rate": 8.722468572550213e-08,
|
| 1797 |
+
"loss": 1.1682,
|
| 1798 |
+
"step": 1250
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 2.609147609147609,
|
| 1802 |
+
"grad_norm": 3.640625,
|
| 1803 |
+
"learning_rate": 8.28264666658851e-08,
|
| 1804 |
+
"loss": 1.1653,
|
| 1805 |
+
"step": 1255
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 2.6195426195426195,
|
| 1809 |
+
"grad_norm": 3.734375,
|
| 1810 |
+
"learning_rate": 7.853723066830486e-08,
|
| 1811 |
+
"loss": 1.1672,
|
| 1812 |
+
"step": 1260
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 2.62993762993763,
|
| 1816 |
+
"grad_norm": 3.640625,
|
| 1817 |
+
"learning_rate": 7.435748740084046e-08,
|
| 1818 |
+
"loss": 1.1606,
|
| 1819 |
+
"step": 1265
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 2.64033264033264,
|
| 1823 |
+
"grad_norm": 3.65625,
|
| 1824 |
+
"learning_rate": 7.028773352110684e-08,
|
| 1825 |
+
"loss": 1.1634,
|
| 1826 |
+
"step": 1270
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 2.6507276507276507,
|
| 1830 |
+
"grad_norm": 3.515625,
|
| 1831 |
+
"learning_rate": 6.632845261724051e-08,
|
| 1832 |
+
"loss": 1.1635,
|
| 1833 |
+
"step": 1275
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 2.6611226611226613,
|
| 1837 |
+
"grad_norm": 4.1875,
|
| 1838 |
+
"learning_rate": 6.248011515043617e-08,
|
| 1839 |
+
"loss": 1.1641,
|
| 1840 |
+
"step": 1280
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 2.6715176715176714,
|
| 1844 |
+
"grad_norm": 3.65625,
|
| 1845 |
+
"learning_rate": 5.8743178399044966e-08,
|
| 1846 |
+
"loss": 1.1642,
|
| 1847 |
+
"step": 1285
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 2.681912681912682,
|
| 1851 |
+
"grad_norm": 3.625,
|
| 1852 |
+
"learning_rate": 5.511808640423765e-08,
|
| 1853 |
+
"loss": 1.1727,
|
| 1854 |
+
"step": 1290
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 2.6923076923076925,
|
| 1858 |
+
"grad_norm": 3.59375,
|
| 1859 |
+
"learning_rate": 5.160526991724246e-08,
|
| 1860 |
+
"loss": 1.1732,
|
| 1861 |
+
"step": 1295
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 2.7027027027027026,
|
| 1865 |
+
"grad_norm": 3.609375,
|
| 1866 |
+
"learning_rate": 4.8205146348160195e-08,
|
| 1867 |
+
"loss": 1.1699,
|
| 1868 |
+
"step": 1300
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 2.713097713097713,
|
| 1872 |
+
"grad_norm": 3.6875,
|
| 1873 |
+
"learning_rate": 4.491811971636605e-08,
|
| 1874 |
+
"loss": 1.166,
|
| 1875 |
+
"step": 1305
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 2.7234927234927238,
|
| 1879 |
+
"grad_norm": 3.625,
|
| 1880 |
+
"learning_rate": 4.174458060250208e-08,
|
| 1881 |
+
"loss": 1.1712,
|
| 1882 |
+
"step": 1310
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 2.733887733887734,
|
| 1886 |
+
"grad_norm": 3.59375,
|
| 1887 |
+
"learning_rate": 3.868490610206565e-08,
|
| 1888 |
+
"loss": 1.1595,
|
| 1889 |
+
"step": 1315
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 2.7442827442827444,
|
| 1893 |
+
"grad_norm": 3.578125,
|
| 1894 |
+
"learning_rate": 3.5739459780602665e-08,
|
| 1895 |
+
"loss": 1.1684,
|
| 1896 |
+
"step": 1320
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 2.7546777546777546,
|
| 1900 |
+
"grad_norm": 3.578125,
|
| 1901 |
+
"learning_rate": 3.290859163050508e-08,
|
| 1902 |
+
"loss": 1.1744,
|
| 1903 |
+
"step": 1325
|
| 1904 |
+
},
|
| 1905 |
+
{
|
| 1906 |
+
"epoch": 2.765072765072765,
|
| 1907 |
+
"grad_norm": 3.671875,
|
| 1908 |
+
"learning_rate": 3.0192638029424735e-08,
|
| 1909 |
+
"loss": 1.1664,
|
| 1910 |
+
"step": 1330
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 2.7754677754677752,
|
| 1914 |
+
"grad_norm": 3.53125,
|
| 1915 |
+
"learning_rate": 2.7591921700302222e-08,
|
| 1916 |
+
"loss": 1.1612,
|
| 1917 |
+
"step": 1335
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"epoch": 2.785862785862786,
|
| 1921 |
+
"grad_norm": 3.609375,
|
| 1922 |
+
"learning_rate": 2.5106751673020012e-08,
|
| 1923 |
+
"loss": 1.174,
|
| 1924 |
+
"step": 1340
|
| 1925 |
+
},
|
| 1926 |
+
{
|
| 1927 |
+
"epoch": 2.7962577962577964,
|
| 1928 |
+
"grad_norm": 3.59375,
|
| 1929 |
+
"learning_rate": 2.273742324768124e-08,
|
| 1930 |
+
"loss": 1.1602,
|
| 1931 |
+
"step": 1345
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 2.8066528066528065,
|
| 1935 |
+
"grad_norm": 3.546875,
|
| 1936 |
+
"learning_rate": 2.048421795952171e-08,
|
| 1937 |
+
"loss": 1.1708,
|
| 1938 |
+
"step": 1350
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 2.817047817047817,
|
| 1942 |
+
"grad_norm": 3.609375,
|
| 1943 |
+
"learning_rate": 1.8347403545455497e-08,
|
| 1944 |
+
"loss": 1.1622,
|
| 1945 |
+
"step": 1355
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 2.8274428274428276,
|
| 1949 |
+
"grad_norm": 3.640625,
|
| 1950 |
+
"learning_rate": 1.6327233912261984e-08,
|
| 1951 |
+
"loss": 1.1668,
|
| 1952 |
+
"step": 1360
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 2.8378378378378377,
|
| 1956 |
+
"grad_norm": 3.609375,
|
| 1957 |
+
"learning_rate": 1.4423949106414868e-08,
|
| 1958 |
+
"loss": 1.1708,
|
| 1959 |
+
"step": 1365
|
| 1960 |
+
},
|
| 1961 |
+
{
|
| 1962 |
+
"epoch": 2.8482328482328483,
|
| 1963 |
+
"grad_norm": 3.640625,
|
| 1964 |
+
"learning_rate": 1.2637775285558983e-08,
|
| 1965 |
+
"loss": 1.1663,
|
| 1966 |
+
"step": 1370
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 2.858627858627859,
|
| 1970 |
+
"grad_norm": 3.5625,
|
| 1971 |
+
"learning_rate": 1.0968924691636572e-08,
|
| 1972 |
+
"loss": 1.1621,
|
| 1973 |
+
"step": 1375
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 2.869022869022869,
|
| 1977 |
+
"grad_norm": 3.671875,
|
| 1978 |
+
"learning_rate": 9.417595625668462e-09,
|
| 1979 |
+
"loss": 1.1769,
|
| 1980 |
+
"step": 1380
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 2.8794178794178795,
|
| 1984 |
+
"grad_norm": 3.71875,
|
| 1985 |
+
"learning_rate": 7.983972424190354e-09,
|
| 1986 |
+
"loss": 1.1784,
|
| 1987 |
+
"step": 1385
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 2.88981288981289,
|
| 1991 |
+
"grad_norm": 3.65625,
|
| 1992 |
+
"learning_rate": 6.668225437349351e-09,
|
| 1993 |
+
"loss": 1.1734,
|
| 1994 |
+
"step": 1390
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 2.9002079002079,
|
| 1998 |
+
"grad_norm": 3.75,
|
| 1999 |
+
"learning_rate": 5.470511008662026e-09,
|
| 2000 |
+
"loss": 1.1747,
|
| 2001 |
+
"step": 1395
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 2.9106029106029108,
|
| 2005 |
+
"grad_norm": 3.671875,
|
| 2006 |
+
"learning_rate": 4.390971456437076e-09,
|
| 2007 |
+
"loss": 1.171,
|
| 2008 |
+
"step": 1400
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 2.920997920997921,
|
| 2012 |
+
"grad_norm": 3.671875,
|
| 2013 |
+
"learning_rate": 3.429735056863725e-09,
|
| 2014 |
+
"loss": 1.1667,
|
| 2015 |
+
"step": 1405
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 2.9313929313929314,
|
| 2019 |
+
"grad_norm": 3.671875,
|
| 2020 |
+
"learning_rate": 2.5869160287702586e-09,
|
| 2021 |
+
"loss": 1.1683,
|
| 2022 |
+
"step": 1410
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 2.9417879417879416,
|
| 2026 |
+
"grad_norm": 3.625,
|
| 2027 |
+
"learning_rate": 1.8626145200513199e-09,
|
| 2028 |
+
"loss": 1.1702,
|
| 2029 |
+
"step": 1415
|
| 2030 |
+
},
|
| 2031 |
+
{
|
| 2032 |
+
"epoch": 2.952182952182952,
|
| 2033 |
+
"grad_norm": 3.5,
|
| 2034 |
+
"learning_rate": 1.2569165957680983e-09,
|
| 2035 |
+
"loss": 1.1712,
|
| 2036 |
+
"step": 1420
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 2.9625779625779627,
|
| 2040 |
+
"grad_norm": 3.640625,
|
| 2041 |
+
"learning_rate": 7.698942279216192e-10,
|
| 2042 |
+
"loss": 1.1725,
|
| 2043 |
+
"step": 1425
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 2.972972972972973,
|
| 2047 |
+
"grad_norm": 3.5,
|
| 2048 |
+
"learning_rate": 4.016052869005859e-10,
|
| 2049 |
+
"loss": 1.1693,
|
| 2050 |
+
"step": 1430
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 2.9833679833679834,
|
| 2054 |
+
"grad_norm": 3.578125,
|
| 2055 |
+
"learning_rate": 1.520935346051022e-10,
|
| 2056 |
+
"loss": 1.1682,
|
| 2057 |
+
"step": 1435
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 2.993762993762994,
|
| 2061 |
+
"grad_norm": 3.609375,
|
| 2062 |
+
"learning_rate": 2.1388619246498486e-11,
|
| 2063 |
+
"loss": 1.1627,
|
| 2064 |
+
"step": 1440
|
| 2065 |
}
|
| 2066 |
],
|
| 2067 |
"logging_steps": 5,
|
|
|
|
| 2076 |
"should_evaluate": false,
|
| 2077 |
"should_log": false,
|
| 2078 |
"should_save": true,
|
| 2079 |
+
"should_training_stop": true
|
| 2080 |
},
|
| 2081 |
"attributes": {}
|
| 2082 |
}
|
| 2083 |
},
|
| 2084 |
+
"total_flos": 8.072899219804914e+18,
|
| 2085 |
"train_batch_size": 8,
|
| 2086 |
"trial_name": null,
|
| 2087 |
"trial_params": null
|