Training in progress, step 7000, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14764744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d12499b0646660a24a6b65eb53983e278236c12bd726c406a28eb8a8cf51f40b
|
| 3 |
size 14764744
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 29614667
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8f752e8418be10a49eba532b536727b787914e487c017b8c91970f6c51d1c73
|
| 3 |
size 29614667
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:799c7c9393a98178f658f197ca2e7f3dcf0c49c3c91d3aaefb6262d7cdbd0b80
|
| 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:c02dce17192fb3005b3efa61adf5ff1250ee4d284cb496a6ef7e19f8df6e311d
|
| 3 |
size 14917
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:103c4217b144607604cf59555b7977070a03537448ea34b568d9fd024a7d6b84
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be9963e9b575d35860b4b24306a82319e3e9ee5be9e379a551ec095ad0d3850d
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 1.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -7334,6 +7334,450 @@
|
|
| 7334 |
"eval_samples_per_second": 5.437,
|
| 7335 |
"eval_steps_per_second": 1.36,
|
| 7336 |
"step": 6600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7337 |
}
|
| 7338 |
],
|
| 7339 |
"logging_steps": 10,
|
|
@@ -7353,7 +7797,7 @@
|
|
| 7353 |
"attributes": {}
|
| 7354 |
}
|
| 7355 |
},
|
| 7356 |
-
"total_flos": 1.
|
| 7357 |
"train_batch_size": 2,
|
| 7358 |
"trial_name": null,
|
| 7359 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.823445131878867,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 7000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 7334 |
"eval_samples_per_second": 5.437,
|
| 7335 |
"eval_steps_per_second": 1.36,
|
| 7336 |
"step": 6600
|
| 7337 |
+
},
|
| 7338 |
+
{
|
| 7339 |
+
"entropy": 0.9823138952255249,
|
| 7340 |
+
"epoch": 1.7218495604037773,
|
| 7341 |
+
"grad_norm": 0.4478851556777954,
|
| 7342 |
+
"learning_rate": 2.7845793175306075e-06,
|
| 7343 |
+
"loss": 0.984638786315918,
|
| 7344 |
+
"mean_token_accuracy": 0.7803937673568726,
|
| 7345 |
+
"num_tokens": 123654.0,
|
| 7346 |
+
"step": 6610
|
| 7347 |
+
},
|
| 7348 |
+
{
|
| 7349 |
+
"entropy": 0.9903601288795472,
|
| 7350 |
+
"epoch": 1.7244545750569849,
|
| 7351 |
+
"grad_norm": 0.4740400016307831,
|
| 7352 |
+
"learning_rate": 2.758530867413389e-06,
|
| 7353 |
+
"loss": 0.9925348281860351,
|
| 7354 |
+
"mean_token_accuracy": 0.7811915218830109,
|
| 7355 |
+
"num_tokens": 250100.0,
|
| 7356 |
+
"step": 6620
|
| 7357 |
+
},
|
| 7358 |
+
{
|
| 7359 |
+
"entropy": 1.0119853645563126,
|
| 7360 |
+
"epoch": 1.7270595897101921,
|
| 7361 |
+
"grad_norm": 0.42794233560562134,
|
| 7362 |
+
"learning_rate": 2.732482417296171e-06,
|
| 7363 |
+
"loss": 1.0036964416503906,
|
| 7364 |
+
"mean_token_accuracy": 0.7763001263141632,
|
| 7365 |
+
"num_tokens": 378633.0,
|
| 7366 |
+
"step": 6630
|
| 7367 |
+
},
|
| 7368 |
+
{
|
| 7369 |
+
"entropy": 0.9493696331977844,
|
| 7370 |
+
"epoch": 1.7296646043633994,
|
| 7371 |
+
"grad_norm": 0.43687593936920166,
|
| 7372 |
+
"learning_rate": 2.706433967178953e-06,
|
| 7373 |
+
"loss": 0.9465467453002929,
|
| 7374 |
+
"mean_token_accuracy": 0.7897346824407577,
|
| 7375 |
+
"num_tokens": 507634.0,
|
| 7376 |
+
"step": 6640
|
| 7377 |
+
},
|
| 7378 |
+
{
|
| 7379 |
+
"entropy": 1.004687362909317,
|
| 7380 |
+
"epoch": 1.732269619016607,
|
| 7381 |
+
"grad_norm": 0.4463178217411041,
|
| 7382 |
+
"learning_rate": 2.680385517061735e-06,
|
| 7383 |
+
"loss": 1.0055112838745117,
|
| 7384 |
+
"mean_token_accuracy": 0.7779219567775726,
|
| 7385 |
+
"num_tokens": 636945.0,
|
| 7386 |
+
"step": 6650
|
| 7387 |
+
},
|
| 7388 |
+
{
|
| 7389 |
+
"entropy": 0.9806973725557327,
|
| 7390 |
+
"epoch": 1.7348746336698144,
|
| 7391 |
+
"grad_norm": 0.4254460334777832,
|
| 7392 |
+
"learning_rate": 2.654337066944517e-06,
|
| 7393 |
+
"loss": 0.9747040748596192,
|
| 7394 |
+
"mean_token_accuracy": 0.7817697480320931,
|
| 7395 |
+
"num_tokens": 766725.0,
|
| 7396 |
+
"step": 6660
|
| 7397 |
+
},
|
| 7398 |
+
{
|
| 7399 |
+
"entropy": 0.9973525032401085,
|
| 7400 |
+
"epoch": 1.737479648323022,
|
| 7401 |
+
"grad_norm": 0.4786764085292816,
|
| 7402 |
+
"learning_rate": 2.6282886168272993e-06,
|
| 7403 |
+
"loss": 1.0005884170532227,
|
| 7404 |
+
"mean_token_accuracy": 0.7777441203594208,
|
| 7405 |
+
"num_tokens": 895306.0,
|
| 7406 |
+
"step": 6670
|
| 7407 |
+
},
|
| 7408 |
+
{
|
| 7409 |
+
"entropy": 0.9907258942723274,
|
| 7410 |
+
"epoch": 1.7400846629762292,
|
| 7411 |
+
"grad_norm": 0.4461391568183899,
|
| 7412 |
+
"learning_rate": 2.602240166710081e-06,
|
| 7413 |
+
"loss": 0.9905254364013671,
|
| 7414 |
+
"mean_token_accuracy": 0.7782276287674904,
|
| 7415 |
+
"num_tokens": 1018780.0,
|
| 7416 |
+
"step": 6680
|
| 7417 |
+
},
|
| 7418 |
+
{
|
| 7419 |
+
"entropy": 0.9964932486414909,
|
| 7420 |
+
"epoch": 1.7426896776294367,
|
| 7421 |
+
"grad_norm": 0.4375130534172058,
|
| 7422 |
+
"learning_rate": 2.576191716592863e-06,
|
| 7423 |
+
"loss": 0.9936812400817872,
|
| 7424 |
+
"mean_token_accuracy": 0.7800995618104934,
|
| 7425 |
+
"num_tokens": 1146973.0,
|
| 7426 |
+
"step": 6690
|
| 7427 |
+
},
|
| 7428 |
+
{
|
| 7429 |
+
"entropy": 0.9998546317219734,
|
| 7430 |
+
"epoch": 1.745294692282644,
|
| 7431 |
+
"grad_norm": 0.5077676177024841,
|
| 7432 |
+
"learning_rate": 2.550143266475645e-06,
|
| 7433 |
+
"loss": 0.9924486160278321,
|
| 7434 |
+
"mean_token_accuracy": 0.7787781774997711,
|
| 7435 |
+
"num_tokens": 1270139.0,
|
| 7436 |
+
"step": 6700
|
| 7437 |
+
},
|
| 7438 |
+
{
|
| 7439 |
+
"epoch": 1.745294692282644,
|
| 7440 |
+
"eval_entropy": 0.988303787839959,
|
| 7441 |
+
"eval_loss": 0.9847902655601501,
|
| 7442 |
+
"eval_mean_token_accuracy": 0.7813749311253271,
|
| 7443 |
+
"eval_num_tokens": 1270139.0,
|
| 7444 |
+
"eval_runtime": 1469.0929,
|
| 7445 |
+
"eval_samples_per_second": 4.633,
|
| 7446 |
+
"eval_steps_per_second": 1.159,
|
| 7447 |
+
"step": 6700
|
| 7448 |
+
},
|
| 7449 |
+
{
|
| 7450 |
+
"entropy": 0.9644292518496513,
|
| 7451 |
+
"epoch": 1.7478997069358515,
|
| 7452 |
+
"grad_norm": 0.49433794617652893,
|
| 7453 |
+
"learning_rate": 2.5240948163584266e-06,
|
| 7454 |
+
"loss": 0.95313720703125,
|
| 7455 |
+
"mean_token_accuracy": 0.7866672441363335,
|
| 7456 |
+
"num_tokens": 1395074.0,
|
| 7457 |
+
"step": 6710
|
| 7458 |
+
},
|
| 7459 |
+
{
|
| 7460 |
+
"entropy": 1.0015375733375549,
|
| 7461 |
+
"epoch": 1.750504721589059,
|
| 7462 |
+
"grad_norm": 0.5286617279052734,
|
| 7463 |
+
"learning_rate": 2.4980463662412086e-06,
|
| 7464 |
+
"loss": 1.0045991897583009,
|
| 7465 |
+
"mean_token_accuracy": 0.7801112249493599,
|
| 7466 |
+
"num_tokens": 1523162.0,
|
| 7467 |
+
"step": 6720
|
| 7468 |
+
},
|
| 7469 |
+
{
|
| 7470 |
+
"entropy": 0.9964019820094109,
|
| 7471 |
+
"epoch": 1.7531097362422665,
|
| 7472 |
+
"grad_norm": 0.5512493252754211,
|
| 7473 |
+
"learning_rate": 2.4719979161239907e-06,
|
| 7474 |
+
"loss": 0.9960369110107422,
|
| 7475 |
+
"mean_token_accuracy": 0.7792898178100586,
|
| 7476 |
+
"num_tokens": 1652179.0,
|
| 7477 |
+
"step": 6730
|
| 7478 |
+
},
|
| 7479 |
+
{
|
| 7480 |
+
"entropy": 0.9651120185852051,
|
| 7481 |
+
"epoch": 1.7557147508954738,
|
| 7482 |
+
"grad_norm": 0.44466137886047363,
|
| 7483 |
+
"learning_rate": 2.4459494660067727e-06,
|
| 7484 |
+
"loss": 0.9617798805236817,
|
| 7485 |
+
"mean_token_accuracy": 0.7847071185708046,
|
| 7486 |
+
"num_tokens": 1780605.0,
|
| 7487 |
+
"step": 6740
|
| 7488 |
+
},
|
| 7489 |
+
{
|
| 7490 |
+
"entropy": 0.9713175401091576,
|
| 7491 |
+
"epoch": 1.758319765548681,
|
| 7492 |
+
"grad_norm": 0.4334625005722046,
|
| 7493 |
+
"learning_rate": 2.4199010158895547e-06,
|
| 7494 |
+
"loss": 0.9693988800048828,
|
| 7495 |
+
"mean_token_accuracy": 0.7834506288170815,
|
| 7496 |
+
"num_tokens": 1907068.0,
|
| 7497 |
+
"step": 6750
|
| 7498 |
+
},
|
| 7499 |
+
{
|
| 7500 |
+
"entropy": 0.9767915159463882,
|
| 7501 |
+
"epoch": 1.7609247802018886,
|
| 7502 |
+
"grad_norm": 0.46460774540901184,
|
| 7503 |
+
"learning_rate": 2.3938525657723368e-06,
|
| 7504 |
+
"loss": 0.9713596343994141,
|
| 7505 |
+
"mean_token_accuracy": 0.783121258020401,
|
| 7506 |
+
"num_tokens": 2036419.0,
|
| 7507 |
+
"step": 6760
|
| 7508 |
+
},
|
| 7509 |
+
{
|
| 7510 |
+
"entropy": 0.9796566858887672,
|
| 7511 |
+
"epoch": 1.763529794855096,
|
| 7512 |
+
"grad_norm": 0.44517257809638977,
|
| 7513 |
+
"learning_rate": 2.367804115655119e-06,
|
| 7514 |
+
"loss": 0.9726932525634766,
|
| 7515 |
+
"mean_token_accuracy": 0.7838401272892952,
|
| 7516 |
+
"num_tokens": 2162634.0,
|
| 7517 |
+
"step": 6770
|
| 7518 |
+
},
|
| 7519 |
+
{
|
| 7520 |
+
"entropy": 1.0208434730768203,
|
| 7521 |
+
"epoch": 1.7661348095083036,
|
| 7522 |
+
"grad_norm": 0.45565611124038696,
|
| 7523 |
+
"learning_rate": 2.341755665537901e-06,
|
| 7524 |
+
"loss": 1.0165723800659179,
|
| 7525 |
+
"mean_token_accuracy": 0.7751889601349831,
|
| 7526 |
+
"num_tokens": 2288676.0,
|
| 7527 |
+
"step": 6780
|
| 7528 |
+
},
|
| 7529 |
+
{
|
| 7530 |
+
"entropy": 0.9510354325175285,
|
| 7531 |
+
"epoch": 1.7687398241615109,
|
| 7532 |
+
"grad_norm": 0.4426063895225525,
|
| 7533 |
+
"learning_rate": 2.315707215420683e-06,
|
| 7534 |
+
"loss": 0.9552410125732422,
|
| 7535 |
+
"mean_token_accuracy": 0.7878434926271438,
|
| 7536 |
+
"num_tokens": 2417023.0,
|
| 7537 |
+
"step": 6790
|
| 7538 |
+
},
|
| 7539 |
+
{
|
| 7540 |
+
"entropy": 1.023421248793602,
|
| 7541 |
+
"epoch": 1.7713448388147184,
|
| 7542 |
+
"grad_norm": 0.45573416352272034,
|
| 7543 |
+
"learning_rate": 2.2896587653034645e-06,
|
| 7544 |
+
"loss": 1.0304821014404297,
|
| 7545 |
+
"mean_token_accuracy": 0.7731378570199012,
|
| 7546 |
+
"num_tokens": 2542335.0,
|
| 7547 |
+
"step": 6800
|
| 7548 |
+
},
|
| 7549 |
+
{
|
| 7550 |
+
"epoch": 1.7713448388147184,
|
| 7551 |
+
"eval_entropy": 0.9875878522875167,
|
| 7552 |
+
"eval_loss": 0.9847276210784912,
|
| 7553 |
+
"eval_mean_token_accuracy": 0.78137964833217,
|
| 7554 |
+
"eval_num_tokens": 2542335.0,
|
| 7555 |
+
"eval_runtime": 1468.1533,
|
| 7556 |
+
"eval_samples_per_second": 4.636,
|
| 7557 |
+
"eval_steps_per_second": 1.159,
|
| 7558 |
+
"step": 6800
|
| 7559 |
+
},
|
| 7560 |
+
{
|
| 7561 |
+
"entropy": 0.9363017573952674,
|
| 7562 |
+
"epoch": 1.7739498534679257,
|
| 7563 |
+
"grad_norm": 0.411294162273407,
|
| 7564 |
+
"learning_rate": 2.2636103151862465e-06,
|
| 7565 |
+
"loss": 0.932148551940918,
|
| 7566 |
+
"mean_token_accuracy": 0.7931686192750931,
|
| 7567 |
+
"num_tokens": 2671515.0,
|
| 7568 |
+
"step": 6810
|
| 7569 |
+
},
|
| 7570 |
+
{
|
| 7571 |
+
"entropy": 1.0269593849778176,
|
| 7572 |
+
"epoch": 1.7765548681211332,
|
| 7573 |
+
"grad_norm": 0.47946611046791077,
|
| 7574 |
+
"learning_rate": 2.2375618650690285e-06,
|
| 7575 |
+
"loss": 1.0208763122558593,
|
| 7576 |
+
"mean_token_accuracy": 0.7766711801290512,
|
| 7577 |
+
"num_tokens": 2796812.0,
|
| 7578 |
+
"step": 6820
|
| 7579 |
+
},
|
| 7580 |
+
{
|
| 7581 |
+
"entropy": 0.9615254417061806,
|
| 7582 |
+
"epoch": 1.7791598827743407,
|
| 7583 |
+
"grad_norm": 0.45359867811203003,
|
| 7584 |
+
"learning_rate": 2.2115134149518106e-06,
|
| 7585 |
+
"loss": 0.9601846694946289,
|
| 7586 |
+
"mean_token_accuracy": 0.7836741924285888,
|
| 7587 |
+
"num_tokens": 2924216.0,
|
| 7588 |
+
"step": 6830
|
| 7589 |
+
},
|
| 7590 |
+
{
|
| 7591 |
+
"entropy": 1.0381957679986953,
|
| 7592 |
+
"epoch": 1.7817648974275482,
|
| 7593 |
+
"grad_norm": 0.4892065227031708,
|
| 7594 |
+
"learning_rate": 2.1854649648345926e-06,
|
| 7595 |
+
"loss": 1.029976749420166,
|
| 7596 |
+
"mean_token_accuracy": 0.7689620569348335,
|
| 7597 |
+
"num_tokens": 3050755.0,
|
| 7598 |
+
"step": 6840
|
| 7599 |
+
},
|
| 7600 |
+
{
|
| 7601 |
+
"entropy": 1.0056249111890794,
|
| 7602 |
+
"epoch": 1.7843699120807555,
|
| 7603 |
+
"grad_norm": 0.5043020844459534,
|
| 7604 |
+
"learning_rate": 2.1594165147173747e-06,
|
| 7605 |
+
"loss": 1.0031335830688477,
|
| 7606 |
+
"mean_token_accuracy": 0.778705857694149,
|
| 7607 |
+
"num_tokens": 3177200.0,
|
| 7608 |
+
"step": 6850
|
| 7609 |
+
},
|
| 7610 |
+
{
|
| 7611 |
+
"entropy": 0.9671053498983383,
|
| 7612 |
+
"epoch": 1.7869749267339627,
|
| 7613 |
+
"grad_norm": 0.4945420026779175,
|
| 7614 |
+
"learning_rate": 2.1333680646001563e-06,
|
| 7615 |
+
"loss": 0.967266845703125,
|
| 7616 |
+
"mean_token_accuracy": 0.7848890349268913,
|
| 7617 |
+
"num_tokens": 3303486.0,
|
| 7618 |
+
"step": 6860
|
| 7619 |
+
},
|
| 7620 |
+
{
|
| 7621 |
+
"entropy": 0.9680422604084015,
|
| 7622 |
+
"epoch": 1.7895799413871702,
|
| 7623 |
+
"grad_norm": 0.4735678732395172,
|
| 7624 |
+
"learning_rate": 2.1073196144829383e-06,
|
| 7625 |
+
"loss": 0.9629220962524414,
|
| 7626 |
+
"mean_token_accuracy": 0.7858682468533515,
|
| 7627 |
+
"num_tokens": 3427886.0,
|
| 7628 |
+
"step": 6870
|
| 7629 |
+
},
|
| 7630 |
+
{
|
| 7631 |
+
"entropy": 0.9950935959815979,
|
| 7632 |
+
"epoch": 1.7921849560403778,
|
| 7633 |
+
"grad_norm": 0.4246147871017456,
|
| 7634 |
+
"learning_rate": 2.0812711643657203e-06,
|
| 7635 |
+
"loss": 0.9934613227844238,
|
| 7636 |
+
"mean_token_accuracy": 0.7778671503067016,
|
| 7637 |
+
"num_tokens": 3558964.0,
|
| 7638 |
+
"step": 6880
|
| 7639 |
+
},
|
| 7640 |
+
{
|
| 7641 |
+
"entropy": 0.9615059599280358,
|
| 7642 |
+
"epoch": 1.7947899706935853,
|
| 7643 |
+
"grad_norm": 0.44503042101860046,
|
| 7644 |
+
"learning_rate": 2.0552227142485024e-06,
|
| 7645 |
+
"loss": 0.9620532989501953,
|
| 7646 |
+
"mean_token_accuracy": 0.7851508602499961,
|
| 7647 |
+
"num_tokens": 3691908.0,
|
| 7648 |
+
"step": 6890
|
| 7649 |
+
},
|
| 7650 |
+
{
|
| 7651 |
+
"entropy": 0.9874796941876411,
|
| 7652 |
+
"epoch": 1.7973949853467925,
|
| 7653 |
+
"grad_norm": 0.4577827751636505,
|
| 7654 |
+
"learning_rate": 2.0291742641312844e-06,
|
| 7655 |
+
"loss": 0.9817806243896484,
|
| 7656 |
+
"mean_token_accuracy": 0.7804968252778053,
|
| 7657 |
+
"num_tokens": 3816343.0,
|
| 7658 |
+
"step": 6900
|
| 7659 |
+
},
|
| 7660 |
+
{
|
| 7661 |
+
"epoch": 1.7973949853467925,
|
| 7662 |
+
"eval_entropy": 0.987803460007969,
|
| 7663 |
+
"eval_loss": 0.9846410155296326,
|
| 7664 |
+
"eval_mean_token_accuracy": 0.7813729251593736,
|
| 7665 |
+
"eval_num_tokens": 3816343.0,
|
| 7666 |
+
"eval_runtime": 1469.6562,
|
| 7667 |
+
"eval_samples_per_second": 4.631,
|
| 7668 |
+
"eval_steps_per_second": 1.158,
|
| 7669 |
+
"step": 6900
|
| 7670 |
+
},
|
| 7671 |
+
{
|
| 7672 |
+
"entropy": 0.9574330061674118,
|
| 7673 |
+
"epoch": 1.8,
|
| 7674 |
+
"grad_norm": 0.44457337260246277,
|
| 7675 |
+
"learning_rate": 2.0031258140140664e-06,
|
| 7676 |
+
"loss": 0.9461252212524414,
|
| 7677 |
+
"mean_token_accuracy": 0.788931405544281,
|
| 7678 |
+
"num_tokens": 3947338.0,
|
| 7679 |
+
"step": 6910
|
| 7680 |
+
},
|
| 7681 |
+
{
|
| 7682 |
+
"entropy": 0.9869294106960297,
|
| 7683 |
+
"epoch": 1.8026050146532073,
|
| 7684 |
+
"grad_norm": 0.46561387181282043,
|
| 7685 |
+
"learning_rate": 1.977077363896848e-06,
|
| 7686 |
+
"loss": 0.983763313293457,
|
| 7687 |
+
"mean_token_accuracy": 0.782055202126503,
|
| 7688 |
+
"num_tokens": 4077287.0,
|
| 7689 |
+
"step": 6920
|
| 7690 |
+
},
|
| 7691 |
+
{
|
| 7692 |
+
"entropy": 1.0384597808122635,
|
| 7693 |
+
"epoch": 1.8052100293064148,
|
| 7694 |
+
"grad_norm": 0.4717126488685608,
|
| 7695 |
+
"learning_rate": 1.9510289137796305e-06,
|
| 7696 |
+
"loss": 1.0406241416931152,
|
| 7697 |
+
"mean_token_accuracy": 0.7699192970991134,
|
| 7698 |
+
"num_tokens": 4202626.0,
|
| 7699 |
+
"step": 6930
|
| 7700 |
+
},
|
| 7701 |
+
{
|
| 7702 |
+
"entropy": 0.9926999747753144,
|
| 7703 |
+
"epoch": 1.8078150439596223,
|
| 7704 |
+
"grad_norm": 0.4420601427555084,
|
| 7705 |
+
"learning_rate": 1.924980463662412e-06,
|
| 7706 |
+
"loss": 0.9950467109680176,
|
| 7707 |
+
"mean_token_accuracy": 0.778462678194046,
|
| 7708 |
+
"num_tokens": 4329768.0,
|
| 7709 |
+
"step": 6940
|
| 7710 |
+
},
|
| 7711 |
+
{
|
| 7712 |
+
"entropy": 1.0370307445526123,
|
| 7713 |
+
"epoch": 1.8104200586128298,
|
| 7714 |
+
"grad_norm": 0.44570279121398926,
|
| 7715 |
+
"learning_rate": 1.8989320135451942e-06,
|
| 7716 |
+
"loss": 1.0290287017822266,
|
| 7717 |
+
"mean_token_accuracy": 0.7743745654821396,
|
| 7718 |
+
"num_tokens": 4454595.0,
|
| 7719 |
+
"step": 6950
|
| 7720 |
+
},
|
| 7721 |
+
{
|
| 7722 |
+
"entropy": 0.9640106305480003,
|
| 7723 |
+
"epoch": 1.8130250732660371,
|
| 7724 |
+
"grad_norm": 0.5364393591880798,
|
| 7725 |
+
"learning_rate": 1.8728835634279762e-06,
|
| 7726 |
+
"loss": 0.9536015510559082,
|
| 7727 |
+
"mean_token_accuracy": 0.786224502325058,
|
| 7728 |
+
"num_tokens": 4579180.0,
|
| 7729 |
+
"step": 6960
|
| 7730 |
+
},
|
| 7731 |
+
{
|
| 7732 |
+
"entropy": 1.0095271036028861,
|
| 7733 |
+
"epoch": 1.8156300879192444,
|
| 7734 |
+
"grad_norm": 0.4075166583061218,
|
| 7735 |
+
"learning_rate": 1.8468351133107582e-06,
|
| 7736 |
+
"loss": 0.9975923538208008,
|
| 7737 |
+
"mean_token_accuracy": 0.7763581544160842,
|
| 7738 |
+
"num_tokens": 4707103.0,
|
| 7739 |
+
"step": 6970
|
| 7740 |
+
},
|
| 7741 |
+
{
|
| 7742 |
+
"entropy": 1.0112810552120208,
|
| 7743 |
+
"epoch": 1.818235102572452,
|
| 7744 |
+
"grad_norm": 0.4442594647407532,
|
| 7745 |
+
"learning_rate": 1.82078666319354e-06,
|
| 7746 |
+
"loss": 1.0171551704406738,
|
| 7747 |
+
"mean_token_accuracy": 0.7736199840903282,
|
| 7748 |
+
"num_tokens": 4835159.0,
|
| 7749 |
+
"step": 6980
|
| 7750 |
+
},
|
| 7751 |
+
{
|
| 7752 |
+
"entropy": 0.9952105522155762,
|
| 7753 |
+
"epoch": 1.8208401172256594,
|
| 7754 |
+
"grad_norm": 0.4370501935482025,
|
| 7755 |
+
"learning_rate": 1.794738213076322e-06,
|
| 7756 |
+
"loss": 0.9866777420043945,
|
| 7757 |
+
"mean_token_accuracy": 0.780173048377037,
|
| 7758 |
+
"num_tokens": 4963278.0,
|
| 7759 |
+
"step": 6990
|
| 7760 |
+
},
|
| 7761 |
+
{
|
| 7762 |
+
"entropy": 1.0098994955420495,
|
| 7763 |
+
"epoch": 1.823445131878867,
|
| 7764 |
+
"grad_norm": 0.4337042272090912,
|
| 7765 |
+
"learning_rate": 1.7686897629591041e-06,
|
| 7766 |
+
"loss": 1.0060459136962892,
|
| 7767 |
+
"mean_token_accuracy": 0.7759813785552978,
|
| 7768 |
+
"num_tokens": 5088254.0,
|
| 7769 |
+
"step": 7000
|
| 7770 |
+
},
|
| 7771 |
+
{
|
| 7772 |
+
"epoch": 1.823445131878867,
|
| 7773 |
+
"eval_entropy": 0.9862276635775135,
|
| 7774 |
+
"eval_loss": 0.984578013420105,
|
| 7775 |
+
"eval_mean_token_accuracy": 0.7813913609950719,
|
| 7776 |
+
"eval_num_tokens": 5088254.0,
|
| 7777 |
+
"eval_runtime": 1468.0381,
|
| 7778 |
+
"eval_samples_per_second": 4.636,
|
| 7779 |
+
"eval_steps_per_second": 1.159,
|
| 7780 |
+
"step": 7000
|
| 7781 |
}
|
| 7782 |
],
|
| 7783 |
"logging_steps": 10,
|
|
|
|
| 7797 |
"attributes": {}
|
| 7798 |
}
|
| 7799 |
},
|
| 7800 |
+
"total_flos": 1.1210055098858209e+18,
|
| 7801 |
"train_batch_size": 2,
|
| 7802 |
"trial_name": null,
|
| 7803 |
"trial_params": null
|