| { | |
| "best_metric": 0.8268348623853211, | |
| "best_model_checkpoint": "tiny-bert-sst2/run-1/checkpoint-1000", | |
| "epoch": 1.8975332068311195, | |
| "eval_steps": 100, | |
| "global_step": 1000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18975332068311196, | |
| "grad_norm": 12.37441349029541, | |
| "learning_rate": 0.0002663036500532049, | |
| "loss": 3.6173, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18975332068311196, | |
| "eval_accuracy": 0.783256880733945, | |
| "eval_loss": 2.335937738418579, | |
| "eval_runtime": 2.5245, | |
| "eval_samples_per_second": 345.413, | |
| "eval_steps_per_second": 2.773, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3795066413662239, | |
| "grad_norm": 27.809226989746094, | |
| "learning_rate": 0.00025130149624925485, | |
| "loss": 1.9873, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3795066413662239, | |
| "eval_accuracy": 0.7889908256880734, | |
| "eval_loss": 2.189484119415283, | |
| "eval_runtime": 2.4106, | |
| "eval_samples_per_second": 361.739, | |
| "eval_steps_per_second": 2.904, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5692599620493358, | |
| "grad_norm": 11.954730033874512, | |
| "learning_rate": 0.00022545996685162219, | |
| "loss": 1.5216, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5692599620493358, | |
| "eval_accuracy": 0.8038990825688074, | |
| "eval_loss": 1.8991937637329102, | |
| "eval_runtime": 2.4589, | |
| "eval_samples_per_second": 354.636, | |
| "eval_steps_per_second": 2.847, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7590132827324478, | |
| "grad_norm": 9.482739448547363, | |
| "learning_rate": 0.00019118302272170003, | |
| "loss": 1.2949, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7590132827324478, | |
| "eval_accuracy": 0.8176605504587156, | |
| "eval_loss": 1.869120717048645, | |
| "eval_runtime": 2.4475, | |
| "eval_samples_per_second": 356.276, | |
| "eval_steps_per_second": 2.86, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9487666034155597, | |
| "grad_norm": 9.123330116271973, | |
| "learning_rate": 0.0001516593463429656, | |
| "loss": 1.1952, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9487666034155597, | |
| "eval_accuracy": 0.8096330275229358, | |
| "eval_loss": 1.8021105527877808, | |
| "eval_runtime": 2.432, | |
| "eval_samples_per_second": 358.555, | |
| "eval_steps_per_second": 2.878, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1385199240986716, | |
| "grad_norm": 11.057866096496582, | |
| "learning_rate": 0.00011056570814378025, | |
| "loss": 0.9849, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1385199240986716, | |
| "eval_accuracy": 0.8211009174311926, | |
| "eval_loss": 1.830503225326538, | |
| "eval_runtime": 2.5073, | |
| "eval_samples_per_second": 347.789, | |
| "eval_steps_per_second": 2.792, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3282732447817835, | |
| "grad_norm": 11.545559883117676, | |
| "learning_rate": 7.172492744582635e-05, | |
| "loss": 0.8491, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.3282732447817835, | |
| "eval_accuracy": 0.823394495412844, | |
| "eval_loss": 1.8194243907928467, | |
| "eval_runtime": 2.4333, | |
| "eval_samples_per_second": 358.362, | |
| "eval_steps_per_second": 2.877, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5180265654648957, | |
| "grad_norm": 12.223103523254395, | |
| "learning_rate": 3.8750246917023896e-05, | |
| "loss": 0.8116, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.5180265654648957, | |
| "eval_accuracy": 0.8188073394495413, | |
| "eval_loss": 1.8401293754577637, | |
| "eval_runtime": 2.4562, | |
| "eval_samples_per_second": 355.021, | |
| "eval_steps_per_second": 2.85, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7077798861480076, | |
| "grad_norm": 15.843563079833984, | |
| "learning_rate": 1.4709203319165587e-05, | |
| "loss": 0.7934, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.7077798861480076, | |
| "eval_accuracy": 0.8268348623853211, | |
| "eval_loss": 1.8587218523025513, | |
| "eval_runtime": 2.4847, | |
| "eval_samples_per_second": 350.954, | |
| "eval_steps_per_second": 2.817, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.8975332068311195, | |
| "grad_norm": 18.251996994018555, | |
| "learning_rate": 1.838263658015889e-06, | |
| "loss": 0.7867, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8975332068311195, | |
| "eval_accuracy": 0.8268348623853211, | |
| "eval_loss": 1.843582034111023, | |
| "eval_runtime": 2.4417, | |
| "eval_samples_per_second": 357.126, | |
| "eval_steps_per_second": 2.867, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1054, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 15348240276300.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.03937708917768734, | |
| "learning_rate": 0.00026956588481964056, | |
| "num_train_epochs": 2, | |
| "temperature": 9, | |
| "warmup_ratio": 0.026036675250872283 | |
| } | |
| } | |