| { | |
| "best_metric": 0.056603044271469116, | |
| "best_model_checkpoint": "autotrain-rp16o-pxwa0/checkpoint-870", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 870, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08620689655172414, | |
| "grad_norm": 3.290644645690918, | |
| "learning_rate": 8.620689655172414e-06, | |
| "loss": 0.3631, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1724137931034483, | |
| "grad_norm": 10.116286277770996, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 0.1219, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.25862068965517243, | |
| "grad_norm": 0.6143497228622437, | |
| "learning_rate": 2.586206896551724e-05, | |
| "loss": 0.1909, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3448275862068966, | |
| "grad_norm": 0.007788954768329859, | |
| "learning_rate": 2.9979600208641352e-05, | |
| "loss": 0.24, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.43103448275862066, | |
| "grad_norm": 6.688470840454102, | |
| "learning_rate": 2.9825994400778473e-05, | |
| "loss": 0.1607, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5172413793103449, | |
| "grad_norm": 6.3299055099487305, | |
| "learning_rate": 2.952334410903845e-05, | |
| "loss": 0.1103, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.603448275862069, | |
| "grad_norm": 0.13505378365516663, | |
| "learning_rate": 2.907469185153564e-05, | |
| "loss": 0.0952, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 6.772790431976318, | |
| "learning_rate": 2.8484547891956387e-05, | |
| "loss": 0.1139, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7758620689655172, | |
| "grad_norm": 2.550690174102783, | |
| "learning_rate": 2.775884489825476e-05, | |
| "loss": 0.1335, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.8620689655172413, | |
| "grad_norm": 11.232149124145508, | |
| "learning_rate": 2.6904878302036937e-05, | |
| "loss": 0.0758, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9482758620689655, | |
| "grad_norm": 1.9099515676498413, | |
| "learning_rate": 2.5931232958196343e-05, | |
| "loss": 0.0902, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cosine_accuracy": 1.0, | |
| "eval_dot_accuracy": 0.0, | |
| "eval_euclidean_accuracy": 1.0, | |
| "eval_loss": 0.06996160000562668, | |
| "eval_manhattan_accuracy": 1.0, | |
| "eval_max_accuracy": 1.0, | |
| "eval_runtime": 44.8595, | |
| "eval_samples_per_second": 12.929, | |
| "eval_steps_per_second": 0.825, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.0344827586206897, | |
| "grad_norm": 0.029423370957374573, | |
| "learning_rate": 2.48476968420842e-05, | |
| "loss": 0.0951, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1206896551724137, | |
| "grad_norm": 0.830912709236145, | |
| "learning_rate": 2.3665162651810512e-05, | |
| "loss": 0.0373, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.206896551724138, | |
| "grad_norm": 8.924799919128418, | |
| "learning_rate": 2.2395518304859387e-05, | |
| "loss": 0.086, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.293103448275862, | |
| "grad_norm": 8.515233039855957, | |
| "learning_rate": 2.105152742984713e-05, | |
| "loss": 0.0418, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "grad_norm": 0.07946328073740005, | |
| "learning_rate": 1.964670105482938e-05, | |
| "loss": 0.0522, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4655172413793103, | |
| "grad_norm": 0.027938440442085266, | |
| "learning_rate": 1.8195161782064143e-05, | |
| "loss": 0.0387, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.5517241379310345, | |
| "grad_norm": 0.20512813329696655, | |
| "learning_rate": 1.6711501814670373e-05, | |
| "loss": 0.0217, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.6379310344827587, | |
| "grad_norm": 0.04595763236284256, | |
| "learning_rate": 1.5210636262428347e-05, | |
| "loss": 0.0455, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 0.002501540817320347, | |
| "learning_rate": 1.3707653201426321e-05, | |
| "loss": 0.0424, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8103448275862069, | |
| "grad_norm": 0.4840604364871979, | |
| "learning_rate": 1.2217661994891308e-05, | |
| "loss": 0.0238, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.896551724137931, | |
| "grad_norm": 13.134848594665527, | |
| "learning_rate": 1.075564140002207e-05, | |
| "loss": 0.0355, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.9827586206896552, | |
| "grad_norm": 1.2007379531860352, | |
| "learning_rate": 9.33628898779359e-06, | |
| "loss": 0.0283, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cosine_accuracy": 1.0, | |
| "eval_dot_accuracy": 0.0, | |
| "eval_euclidean_accuracy": 1.0, | |
| "eval_loss": 0.059745438396930695, | |
| "eval_manhattan_accuracy": 1.0, | |
| "eval_max_accuracy": 1.0, | |
| "eval_runtime": 45.055, | |
| "eval_samples_per_second": 12.873, | |
| "eval_steps_per_second": 0.821, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.0689655172413794, | |
| "grad_norm": 0.010960499756038189, | |
| "learning_rate": 7.97387338950315e-06, | |
| "loss": 0.0213, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.1551724137931036, | |
| "grad_norm": 0.004453401546925306, | |
| "learning_rate": 6.682090855411221e-06, | |
| "loss": 0.0219, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.2413793103448274, | |
| "grad_norm": 7.022181987762451, | |
| "learning_rate": 5.473927567481096e-06, | |
| "loss": 0.0254, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.3275862068965516, | |
| "grad_norm": 1.2289260625839233, | |
| "learning_rate": 4.361529090375834e-06, | |
| "loss": 0.0204, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.413793103448276, | |
| "grad_norm": 0.006588762626051903, | |
| "learning_rate": 3.35607827311076e-06, | |
| "loss": 0.0052, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.5081865787506104, | |
| "learning_rate": 2.467682828805956e-06, | |
| "loss": 0.0248, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.586206896551724, | |
| "grad_norm": 17.075759887695312, | |
| "learning_rate": 1.7052737226901876e-06, | |
| "loss": 0.0507, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.6724137931034484, | |
| "grad_norm": 0.9192395806312561, | |
| "learning_rate": 1.0765153898531083e-06, | |
| "loss": 0.0191, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.7586206896551726, | |
| "grad_norm": 0.03410585597157478, | |
| "learning_rate": 5.877286853191999e-07, | |
| "loss": 0.018, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.844827586206897, | |
| "grad_norm": 0.09378518909215927, | |
| "learning_rate": 2.438273410199598e-07, | |
| "loss": 0.0176, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.9310344827586206, | |
| "grad_norm": 0.024510715156793594, | |
| "learning_rate": 4.826856845703165e-08, | |
| "loss": 0.0193, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cosine_accuracy": 1.0, | |
| "eval_dot_accuracy": 0.0, | |
| "eval_euclidean_accuracy": 1.0, | |
| "eval_loss": 0.056603044271469116, | |
| "eval_manhattan_accuracy": 1.0, | |
| "eval_max_accuracy": 1.0, | |
| "eval_runtime": 43.366, | |
| "eval_samples_per_second": 13.375, | |
| "eval_steps_per_second": 0.853, | |
| "step": 870 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 870, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |