| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9962546816479403, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.049937578027465665, | |
| "grad_norm": 29.759422123901825, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0298, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09987515605493133, | |
| "grad_norm": 1.659620518169597, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9418, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.149812734082397, | |
| "grad_norm": 1.6655817038532692, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9067, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19975031210986266, | |
| "grad_norm": 3.758532618440977, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8837, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24968789013732834, | |
| "grad_norm": 1.0914089528556838, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8683, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.299625468164794, | |
| "grad_norm": 0.8217188574851603, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8554, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3495630461922597, | |
| "grad_norm": 2.6152137256768913, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8501, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3995006242197253, | |
| "grad_norm": 0.6790645381435163, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8376, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.449438202247191, | |
| "grad_norm": 1.050547891408586, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8389, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4993757802746567, | |
| "grad_norm": 0.61531936130051, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8283, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5493133583021224, | |
| "grad_norm": 0.7158756598167714, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8285, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.599250936329588, | |
| "grad_norm": 0.6741740850175894, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8208, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6491885143570537, | |
| "grad_norm": 0.727395321147856, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8226, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6991260923845194, | |
| "grad_norm": 0.5979029940330495, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8218, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7490636704119851, | |
| "grad_norm": 0.7354628133399658, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8146, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7990012484394506, | |
| "grad_norm": 1.0797869029852383, | |
| "learning_rate": 5e-06, | |
| "loss": 0.811, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8489388264669163, | |
| "grad_norm": 0.5732338176348897, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8117, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.898876404494382, | |
| "grad_norm": 0.516647597175339, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8099, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9488139825218477, | |
| "grad_norm": 0.595235164677505, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8062, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9987515605493134, | |
| "grad_norm": 0.5491264032653016, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7992, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9987515605493134, | |
| "eval_loss": 0.8015628457069397, | |
| "eval_runtime": 212.5855, | |
| "eval_samples_per_second": 25.387, | |
| "eval_steps_per_second": 0.4, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.048689138576779, | |
| "grad_norm": 1.1815779977106664, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8245, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0986267166042447, | |
| "grad_norm": 0.8607004375758024, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7537, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.1485642946317103, | |
| "grad_norm": 0.6382531480247193, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7617, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.198501872659176, | |
| "grad_norm": 0.5848998545511357, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7635, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.2484394506866416, | |
| "grad_norm": 0.6856479277932508, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7623, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2983770287141074, | |
| "grad_norm": 0.7819524787327043, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7567, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.348314606741573, | |
| "grad_norm": 0.6831313099201878, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7606, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.3982521847690386, | |
| "grad_norm": 0.6700856388131974, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7553, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.4481897627965044, | |
| "grad_norm": 0.5874295240823044, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7575, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.4981273408239701, | |
| "grad_norm": 0.6100148315517313, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7523, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5480649188514357, | |
| "grad_norm": 0.6291672713518774, | |
| "learning_rate": 5e-06, | |
| "loss": 0.759, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.5980024968789013, | |
| "grad_norm": 0.7275448418797654, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7532, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.647940074906367, | |
| "grad_norm": 0.6798292981346045, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7652, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.6978776529338329, | |
| "grad_norm": 0.7320780258400261, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7554, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.7478152309612984, | |
| "grad_norm": 0.6107676047027211, | |
| "learning_rate": 5e-06, | |
| "loss": 0.757, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.797752808988764, | |
| "grad_norm": 0.5083613384732135, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7576, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.8476903870162298, | |
| "grad_norm": 0.5021025632111004, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7584, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.8976279650436954, | |
| "grad_norm": 0.6593302140861815, | |
| "learning_rate": 5e-06, | |
| "loss": 0.755, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.947565543071161, | |
| "grad_norm": 0.5669868340257436, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7478, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.9975031210986267, | |
| "grad_norm": 0.5745040341281294, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7538, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.9975031210986267, | |
| "eval_loss": 0.7874204516410828, | |
| "eval_runtime": 212.6782, | |
| "eval_samples_per_second": 25.376, | |
| "eval_steps_per_second": 0.4, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0474406991260925, | |
| "grad_norm": 0.6952347883899184, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7611, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.097378277153558, | |
| "grad_norm": 0.6000479994459602, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7037, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.1473158551810236, | |
| "grad_norm": 0.6575266096005482, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7089, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.1972534332084894, | |
| "grad_norm": 0.7384159721059136, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7057, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.247191011235955, | |
| "grad_norm": 0.6735840214535883, | |
| "learning_rate": 5e-06, | |
| "loss": 0.706, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2971285892634206, | |
| "grad_norm": 0.7153617513297972, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7064, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.3470661672908864, | |
| "grad_norm": 0.7396133098853745, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7049, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.397003745318352, | |
| "grad_norm": 0.6440383221784979, | |
| "learning_rate": 5e-06, | |
| "loss": 0.705, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.4469413233458175, | |
| "grad_norm": 0.5481603423583875, | |
| "learning_rate": 5e-06, | |
| "loss": 0.709, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.4968789013732833, | |
| "grad_norm": 0.6611697985224058, | |
| "learning_rate": 5e-06, | |
| "loss": 0.71, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.546816479400749, | |
| "grad_norm": 0.6252639550455323, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7128, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.596754057428215, | |
| "grad_norm": 0.578764019014536, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7116, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.6466916354556806, | |
| "grad_norm": 0.5718219886250622, | |
| "learning_rate": 5e-06, | |
| "loss": 0.711, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.696629213483146, | |
| "grad_norm": 0.6480951015929783, | |
| "learning_rate": 5e-06, | |
| "loss": 0.706, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.746566791510612, | |
| "grad_norm": 0.568128147930456, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7144, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.796504369538077, | |
| "grad_norm": 0.7016907742592169, | |
| "learning_rate": 5e-06, | |
| "loss": 0.708, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.846441947565543, | |
| "grad_norm": 0.6718047517989062, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7147, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.8963795255930087, | |
| "grad_norm": 0.7869266984488797, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7114, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.9463171036204745, | |
| "grad_norm": 0.6070316014377024, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7072, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.9962546816479403, | |
| "grad_norm": 0.5662761836861052, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7158, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9962546816479403, | |
| "eval_loss": 0.7859531044960022, | |
| "eval_runtime": 213.866, | |
| "eval_samples_per_second": 25.235, | |
| "eval_steps_per_second": 0.397, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9962546816479403, | |
| "step": 600, | |
| "total_flos": 1004812967608320.0, | |
| "train_loss": 0.773849273522695, | |
| "train_runtime": 35442.7976, | |
| "train_samples_per_second": 8.678, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1004812967608320.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |