Training checkpoint at step 27000
Browse files- trainer_state.json +365 -5
trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 2.
|
| 4 |
"best_model_checkpoint": "../runs/karpathy/fineweb-edu-100b-shuffle/meta-llama/Llama-3.2-1B/linear_adamw_wd1e-03_8x1024_mem32_bs64_hf_armt_dmem64/run_21/checkpoint-25000",
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -9368,6 +9368,366 @@
|
|
| 9368 |
"eval_samples_per_second": 3.212,
|
| 9369 |
"eval_steps_per_second": 1.606,
|
| 9370 |
"step": 26000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9371 |
}
|
| 9372 |
],
|
| 9373 |
"logging_steps": 25,
|
|
@@ -9387,7 +9747,7 @@
|
|
| 9387 |
"attributes": {}
|
| 9388 |
}
|
| 9389 |
},
|
| 9390 |
-
"total_flos": 8.
|
| 9391 |
"train_batch_size": 1,
|
| 9392 |
"trial_name": null,
|
| 9393 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 26800,
|
| 3 |
+
"best_metric": 2.381396532058716,
|
| 4 |
"best_model_checkpoint": "../runs/karpathy/fineweb-edu-100b-shuffle/meta-llama/Llama-3.2-1B/linear_adamw_wd1e-03_8x1024_mem32_bs64_hf_armt_dmem64/run_21/checkpoint-25000",
|
| 5 |
+
"epoch": 0.54,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 27000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 9368 |
"eval_samples_per_second": 3.212,
|
| 9369 |
"eval_steps_per_second": 1.606,
|
| 9370 |
"step": 26000
|
| 9371 |
+
},
|
| 9372 |
+
{
|
| 9373 |
+
"epoch": 0.5205,
|
| 9374 |
+
"grad_norm": 0.5425846904290578,
|
| 9375 |
+
"learning_rate": 5.328000000000001e-06,
|
| 9376 |
+
"loss": 2.3675,
|
| 9377 |
+
"step": 26025
|
| 9378 |
+
},
|
| 9379 |
+
{
|
| 9380 |
+
"epoch": 0.521,
|
| 9381 |
+
"grad_norm": 0.5621800567569987,
|
| 9382 |
+
"learning_rate": 5.322444444444445e-06,
|
| 9383 |
+
"loss": 2.3759,
|
| 9384 |
+
"step": 26050
|
| 9385 |
+
},
|
| 9386 |
+
{
|
| 9387 |
+
"epoch": 0.5215,
|
| 9388 |
+
"grad_norm": 0.5544103291449336,
|
| 9389 |
+
"learning_rate": 5.316888888888889e-06,
|
| 9390 |
+
"loss": 2.3576,
|
| 9391 |
+
"step": 26075
|
| 9392 |
+
},
|
| 9393 |
+
{
|
| 9394 |
+
"epoch": 0.522,
|
| 9395 |
+
"grad_norm": 0.550125457461572,
|
| 9396 |
+
"learning_rate": 5.311333333333334e-06,
|
| 9397 |
+
"loss": 2.3567,
|
| 9398 |
+
"step": 26100
|
| 9399 |
+
},
|
| 9400 |
+
{
|
| 9401 |
+
"epoch": 0.522,
|
| 9402 |
+
"eval_loss": 2.382749319076538,
|
| 9403 |
+
"eval_runtime": 31.8184,
|
| 9404 |
+
"eval_samples_per_second": 3.206,
|
| 9405 |
+
"eval_steps_per_second": 1.603,
|
| 9406 |
+
"step": 26100
|
| 9407 |
+
},
|
| 9408 |
+
{
|
| 9409 |
+
"epoch": 0.5225,
|
| 9410 |
+
"grad_norm": 0.5441956885780074,
|
| 9411 |
+
"learning_rate": 5.305777777777779e-06,
|
| 9412 |
+
"loss": 2.3562,
|
| 9413 |
+
"step": 26125
|
| 9414 |
+
},
|
| 9415 |
+
{
|
| 9416 |
+
"epoch": 0.523,
|
| 9417 |
+
"grad_norm": 0.5677266247403775,
|
| 9418 |
+
"learning_rate": 5.3002222222222225e-06,
|
| 9419 |
+
"loss": 2.3666,
|
| 9420 |
+
"step": 26150
|
| 9421 |
+
},
|
| 9422 |
+
{
|
| 9423 |
+
"epoch": 0.5235,
|
| 9424 |
+
"grad_norm": 0.5396975563673215,
|
| 9425 |
+
"learning_rate": 5.294666666666667e-06,
|
| 9426 |
+
"loss": 2.351,
|
| 9427 |
+
"step": 26175
|
| 9428 |
+
},
|
| 9429 |
+
{
|
| 9430 |
+
"epoch": 0.524,
|
| 9431 |
+
"grad_norm": 0.5374437057610971,
|
| 9432 |
+
"learning_rate": 5.289111111111111e-06,
|
| 9433 |
+
"loss": 2.3625,
|
| 9434 |
+
"step": 26200
|
| 9435 |
+
},
|
| 9436 |
+
{
|
| 9437 |
+
"epoch": 0.524,
|
| 9438 |
+
"eval_loss": 2.3822991847991943,
|
| 9439 |
+
"eval_runtime": 31.8822,
|
| 9440 |
+
"eval_samples_per_second": 3.199,
|
| 9441 |
+
"eval_steps_per_second": 1.6,
|
| 9442 |
+
"step": 26200
|
| 9443 |
+
},
|
| 9444 |
+
{
|
| 9445 |
+
"epoch": 0.5245,
|
| 9446 |
+
"grad_norm": 0.5627076715491244,
|
| 9447 |
+
"learning_rate": 5.2835555555555566e-06,
|
| 9448 |
+
"loss": 2.3699,
|
| 9449 |
+
"step": 26225
|
| 9450 |
+
},
|
| 9451 |
+
{
|
| 9452 |
+
"epoch": 0.525,
|
| 9453 |
+
"grad_norm": 0.5430691314388109,
|
| 9454 |
+
"learning_rate": 5.278000000000001e-06,
|
| 9455 |
+
"loss": 2.3648,
|
| 9456 |
+
"step": 26250
|
| 9457 |
+
},
|
| 9458 |
+
{
|
| 9459 |
+
"epoch": 0.5255,
|
| 9460 |
+
"grad_norm": 0.5319128139639624,
|
| 9461 |
+
"learning_rate": 5.2724444444444445e-06,
|
| 9462 |
+
"loss": 2.3722,
|
| 9463 |
+
"step": 26275
|
| 9464 |
+
},
|
| 9465 |
+
{
|
| 9466 |
+
"epoch": 0.526,
|
| 9467 |
+
"grad_norm": 0.5560009569047116,
|
| 9468 |
+
"learning_rate": 5.266888888888889e-06,
|
| 9469 |
+
"loss": 2.3763,
|
| 9470 |
+
"step": 26300
|
| 9471 |
+
},
|
| 9472 |
+
{
|
| 9473 |
+
"epoch": 0.526,
|
| 9474 |
+
"eval_loss": 2.3822247982025146,
|
| 9475 |
+
"eval_runtime": 31.7558,
|
| 9476 |
+
"eval_samples_per_second": 3.212,
|
| 9477 |
+
"eval_steps_per_second": 1.606,
|
| 9478 |
+
"step": 26300
|
| 9479 |
+
},
|
| 9480 |
+
{
|
| 9481 |
+
"epoch": 0.5265,
|
| 9482 |
+
"grad_norm": 0.5586923319248112,
|
| 9483 |
+
"learning_rate": 5.261333333333334e-06,
|
| 9484 |
+
"loss": 2.366,
|
| 9485 |
+
"step": 26325
|
| 9486 |
+
},
|
| 9487 |
+
{
|
| 9488 |
+
"epoch": 0.527,
|
| 9489 |
+
"grad_norm": 0.5621950392943218,
|
| 9490 |
+
"learning_rate": 5.255777777777779e-06,
|
| 9491 |
+
"loss": 2.3713,
|
| 9492 |
+
"step": 26350
|
| 9493 |
+
},
|
| 9494 |
+
{
|
| 9495 |
+
"epoch": 0.5275,
|
| 9496 |
+
"grad_norm": 0.5630783729958978,
|
| 9497 |
+
"learning_rate": 5.250222222222222e-06,
|
| 9498 |
+
"loss": 2.3508,
|
| 9499 |
+
"step": 26375
|
| 9500 |
+
},
|
| 9501 |
+
{
|
| 9502 |
+
"epoch": 0.528,
|
| 9503 |
+
"grad_norm": 0.5543463911581646,
|
| 9504 |
+
"learning_rate": 5.2446666666666665e-06,
|
| 9505 |
+
"loss": 2.3588,
|
| 9506 |
+
"step": 26400
|
| 9507 |
+
},
|
| 9508 |
+
{
|
| 9509 |
+
"epoch": 0.528,
|
| 9510 |
+
"eval_loss": 2.3820412158966064,
|
| 9511 |
+
"eval_runtime": 31.7735,
|
| 9512 |
+
"eval_samples_per_second": 3.21,
|
| 9513 |
+
"eval_steps_per_second": 1.605,
|
| 9514 |
+
"step": 26400
|
| 9515 |
+
},
|
| 9516 |
+
{
|
| 9517 |
+
"epoch": 0.5285,
|
| 9518 |
+
"grad_norm": 0.5521701819516005,
|
| 9519 |
+
"learning_rate": 5.239111111111112e-06,
|
| 9520 |
+
"loss": 2.3798,
|
| 9521 |
+
"step": 26425
|
| 9522 |
+
},
|
| 9523 |
+
{
|
| 9524 |
+
"epoch": 0.529,
|
| 9525 |
+
"grad_norm": 0.5697290541696707,
|
| 9526 |
+
"learning_rate": 5.233555555555556e-06,
|
| 9527 |
+
"loss": 2.3705,
|
| 9528 |
+
"step": 26450
|
| 9529 |
+
},
|
| 9530 |
+
{
|
| 9531 |
+
"epoch": 0.5295,
|
| 9532 |
+
"grad_norm": 0.5456656767494042,
|
| 9533 |
+
"learning_rate": 5.228000000000001e-06,
|
| 9534 |
+
"loss": 2.3603,
|
| 9535 |
+
"step": 26475
|
| 9536 |
+
},
|
| 9537 |
+
{
|
| 9538 |
+
"epoch": 0.53,
|
| 9539 |
+
"grad_norm": 0.544157308823069,
|
| 9540 |
+
"learning_rate": 5.222444444444444e-06,
|
| 9541 |
+
"loss": 2.3598,
|
| 9542 |
+
"step": 26500
|
| 9543 |
+
},
|
| 9544 |
+
{
|
| 9545 |
+
"epoch": 0.53,
|
| 9546 |
+
"eval_loss": 2.3819408416748047,
|
| 9547 |
+
"eval_runtime": 31.804,
|
| 9548 |
+
"eval_samples_per_second": 3.207,
|
| 9549 |
+
"eval_steps_per_second": 1.604,
|
| 9550 |
+
"step": 26500
|
| 9551 |
+
},
|
| 9552 |
+
{
|
| 9553 |
+
"epoch": 0.5305,
|
| 9554 |
+
"grad_norm": 0.5399718074412095,
|
| 9555 |
+
"learning_rate": 5.216888888888889e-06,
|
| 9556 |
+
"loss": 2.3765,
|
| 9557 |
+
"step": 26525
|
| 9558 |
+
},
|
| 9559 |
+
{
|
| 9560 |
+
"epoch": 0.531,
|
| 9561 |
+
"grad_norm": 0.542440216852853,
|
| 9562 |
+
"learning_rate": 5.211333333333334e-06,
|
| 9563 |
+
"loss": 2.3758,
|
| 9564 |
+
"step": 26550
|
| 9565 |
+
},
|
| 9566 |
+
{
|
| 9567 |
+
"epoch": 0.5315,
|
| 9568 |
+
"grad_norm": 0.5648571300651365,
|
| 9569 |
+
"learning_rate": 5.205777777777778e-06,
|
| 9570 |
+
"loss": 2.3685,
|
| 9571 |
+
"step": 26575
|
| 9572 |
+
},
|
| 9573 |
+
{
|
| 9574 |
+
"epoch": 0.532,
|
| 9575 |
+
"grad_norm": 0.573442767423831,
|
| 9576 |
+
"learning_rate": 5.2002222222222235e-06,
|
| 9577 |
+
"loss": 2.3556,
|
| 9578 |
+
"step": 26600
|
| 9579 |
+
},
|
| 9580 |
+
{
|
| 9581 |
+
"epoch": 0.532,
|
| 9582 |
+
"eval_loss": 2.382056951522827,
|
| 9583 |
+
"eval_runtime": 31.8038,
|
| 9584 |
+
"eval_samples_per_second": 3.207,
|
| 9585 |
+
"eval_steps_per_second": 1.604,
|
| 9586 |
+
"step": 26600
|
| 9587 |
+
},
|
| 9588 |
+
{
|
| 9589 |
+
"epoch": 0.5325,
|
| 9590 |
+
"grad_norm": 0.6056414806190663,
|
| 9591 |
+
"learning_rate": 5.194666666666667e-06,
|
| 9592 |
+
"loss": 2.3595,
|
| 9593 |
+
"step": 26625
|
| 9594 |
+
},
|
| 9595 |
+
{
|
| 9596 |
+
"epoch": 0.533,
|
| 9597 |
+
"grad_norm": 0.5481757619700885,
|
| 9598 |
+
"learning_rate": 5.189111111111111e-06,
|
| 9599 |
+
"loss": 2.3727,
|
| 9600 |
+
"step": 26650
|
| 9601 |
+
},
|
| 9602 |
+
{
|
| 9603 |
+
"epoch": 0.5335,
|
| 9604 |
+
"grad_norm": 0.5610562792027696,
|
| 9605 |
+
"learning_rate": 5.183555555555556e-06,
|
| 9606 |
+
"loss": 2.3673,
|
| 9607 |
+
"step": 26675
|
| 9608 |
+
},
|
| 9609 |
+
{
|
| 9610 |
+
"epoch": 0.534,
|
| 9611 |
+
"grad_norm": 0.5702347426339772,
|
| 9612 |
+
"learning_rate": 5.178000000000001e-06,
|
| 9613 |
+
"loss": 2.3622,
|
| 9614 |
+
"step": 26700
|
| 9615 |
+
},
|
| 9616 |
+
{
|
| 9617 |
+
"epoch": 0.534,
|
| 9618 |
+
"eval_loss": 2.381828546524048,
|
| 9619 |
+
"eval_runtime": 31.992,
|
| 9620 |
+
"eval_samples_per_second": 3.188,
|
| 9621 |
+
"eval_steps_per_second": 1.594,
|
| 9622 |
+
"step": 26700
|
| 9623 |
+
},
|
| 9624 |
+
{
|
| 9625 |
+
"epoch": 0.5345,
|
| 9626 |
+
"grad_norm": 0.5565593579595437,
|
| 9627 |
+
"learning_rate": 5.172444444444445e-06,
|
| 9628 |
+
"loss": 2.3651,
|
| 9629 |
+
"step": 26725
|
| 9630 |
+
},
|
| 9631 |
+
{
|
| 9632 |
+
"epoch": 0.535,
|
| 9633 |
+
"grad_norm": 0.5398272748687973,
|
| 9634 |
+
"learning_rate": 5.166888888888889e-06,
|
| 9635 |
+
"loss": 2.3703,
|
| 9636 |
+
"step": 26750
|
| 9637 |
+
},
|
| 9638 |
+
{
|
| 9639 |
+
"epoch": 0.5355,
|
| 9640 |
+
"grad_norm": 0.5611538131409728,
|
| 9641 |
+
"learning_rate": 5.1613333333333334e-06,
|
| 9642 |
+
"loss": 2.3778,
|
| 9643 |
+
"step": 26775
|
| 9644 |
+
},
|
| 9645 |
+
{
|
| 9646 |
+
"epoch": 0.536,
|
| 9647 |
+
"grad_norm": 0.5436520053621182,
|
| 9648 |
+
"learning_rate": 5.155777777777779e-06,
|
| 9649 |
+
"loss": 2.3561,
|
| 9650 |
+
"step": 26800
|
| 9651 |
+
},
|
| 9652 |
+
{
|
| 9653 |
+
"epoch": 0.536,
|
| 9654 |
+
"eval_loss": 2.381396532058716,
|
| 9655 |
+
"eval_runtime": 31.772,
|
| 9656 |
+
"eval_samples_per_second": 3.21,
|
| 9657 |
+
"eval_steps_per_second": 1.605,
|
| 9658 |
+
"step": 26800
|
| 9659 |
+
},
|
| 9660 |
+
{
|
| 9661 |
+
"epoch": 0.5365,
|
| 9662 |
+
"grad_norm": 0.5574841239488896,
|
| 9663 |
+
"learning_rate": 5.150222222222223e-06,
|
| 9664 |
+
"loss": 2.3607,
|
| 9665 |
+
"step": 26825
|
| 9666 |
+
},
|
| 9667 |
+
{
|
| 9668 |
+
"epoch": 0.537,
|
| 9669 |
+
"grad_norm": 0.5459267231396281,
|
| 9670 |
+
"learning_rate": 5.144666666666667e-06,
|
| 9671 |
+
"loss": 2.3652,
|
| 9672 |
+
"step": 26850
|
| 9673 |
+
},
|
| 9674 |
+
{
|
| 9675 |
+
"epoch": 0.5375,
|
| 9676 |
+
"grad_norm": 0.5764624554311072,
|
| 9677 |
+
"learning_rate": 5.139111111111111e-06,
|
| 9678 |
+
"loss": 2.3748,
|
| 9679 |
+
"step": 26875
|
| 9680 |
+
},
|
| 9681 |
+
{
|
| 9682 |
+
"epoch": 0.538,
|
| 9683 |
+
"grad_norm": 0.5452582655691465,
|
| 9684 |
+
"learning_rate": 5.133555555555556e-06,
|
| 9685 |
+
"loss": 2.3751,
|
| 9686 |
+
"step": 26900
|
| 9687 |
+
},
|
| 9688 |
+
{
|
| 9689 |
+
"epoch": 0.538,
|
| 9690 |
+
"eval_loss": 2.3815813064575195,
|
| 9691 |
+
"eval_runtime": 31.833,
|
| 9692 |
+
"eval_samples_per_second": 3.204,
|
| 9693 |
+
"eval_steps_per_second": 1.602,
|
| 9694 |
+
"step": 26900
|
| 9695 |
+
},
|
| 9696 |
+
{
|
| 9697 |
+
"epoch": 0.5385,
|
| 9698 |
+
"grad_norm": 0.5591974032204698,
|
| 9699 |
+
"learning_rate": 5.128000000000001e-06,
|
| 9700 |
+
"loss": 2.3595,
|
| 9701 |
+
"step": 26925
|
| 9702 |
+
},
|
| 9703 |
+
{
|
| 9704 |
+
"epoch": 0.539,
|
| 9705 |
+
"grad_norm": 0.5910956937930101,
|
| 9706 |
+
"learning_rate": 5.122444444444444e-06,
|
| 9707 |
+
"loss": 2.3712,
|
| 9708 |
+
"step": 26950
|
| 9709 |
+
},
|
| 9710 |
+
{
|
| 9711 |
+
"epoch": 0.5395,
|
| 9712 |
+
"grad_norm": 0.5532516136915937,
|
| 9713 |
+
"learning_rate": 5.116888888888889e-06,
|
| 9714 |
+
"loss": 2.3673,
|
| 9715 |
+
"step": 26975
|
| 9716 |
+
},
|
| 9717 |
+
{
|
| 9718 |
+
"epoch": 0.54,
|
| 9719 |
+
"grad_norm": 0.5654498740726267,
|
| 9720 |
+
"learning_rate": 5.111333333333334e-06,
|
| 9721 |
+
"loss": 2.3667,
|
| 9722 |
+
"step": 27000
|
| 9723 |
+
},
|
| 9724 |
+
{
|
| 9725 |
+
"epoch": 0.54,
|
| 9726 |
+
"eval_loss": 2.3814122676849365,
|
| 9727 |
+
"eval_runtime": 31.7588,
|
| 9728 |
+
"eval_samples_per_second": 3.212,
|
| 9729 |
+
"eval_steps_per_second": 1.606,
|
| 9730 |
+
"step": 27000
|
| 9731 |
}
|
| 9732 |
],
|
| 9733 |
"logging_steps": 25,
|
|
|
|
| 9747 |
"attributes": {}
|
| 9748 |
}
|
| 9749 |
},
|
| 9750 |
+
"total_flos": 8.59464620548345e+19,
|
| 9751 |
"train_batch_size": 1,
|
| 9752 |
"trial_name": null,
|
| 9753 |
"trial_params": null
|