| { | |
| "best_metric": 0.6848593354225159, | |
| "best_model_checkpoint": "bert_base_lda_5_wnli/checkpoint-30", | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 45, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.482614517211914, | |
| "learning_rate": 0.0009666666666666667, | |
| "loss": 1.121, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.820504903793335, | |
| "eval_runtime": 0.0522, | |
| "eval_samples_per_second": 1359.977, | |
| "eval_steps_per_second": 19.155, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 15.61330509185791, | |
| "learning_rate": 0.0009333333333333333, | |
| "loss": 1.6034, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 1.7292718887329102, | |
| "eval_runtime": 0.0552, | |
| "eval_samples_per_second": 1286.613, | |
| "eval_steps_per_second": 18.121, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.5773879885673523, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.9483, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.764909565448761, | |
| "eval_runtime": 0.0521, | |
| "eval_samples_per_second": 1364.026, | |
| "eval_steps_per_second": 19.212, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.563255786895752, | |
| "learning_rate": 0.0008666666666666667, | |
| "loss": 0.7514, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.7557218074798584, | |
| "eval_runtime": 0.0521, | |
| "eval_samples_per_second": 1362.466, | |
| "eval_steps_per_second": 19.19, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.8208609223365784, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 0.746, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.8104642033576965, | |
| "eval_runtime": 0.052, | |
| "eval_samples_per_second": 1366.58, | |
| "eval_steps_per_second": 19.248, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.4645087718963623, | |
| "learning_rate": 0.0008, | |
| "loss": 0.7896, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.7383360266685486, | |
| "eval_runtime": 0.0519, | |
| "eval_samples_per_second": 1366.75, | |
| "eval_steps_per_second": 19.25, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.9648402333259583, | |
| "learning_rate": 0.0007666666666666667, | |
| "loss": 0.7573, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.685299277305603, | |
| "eval_runtime": 0.053, | |
| "eval_samples_per_second": 1338.925, | |
| "eval_steps_per_second": 18.858, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.858669102191925, | |
| "learning_rate": 0.0007333333333333333, | |
| "loss": 0.6951, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.8345620036125183, | |
| "eval_runtime": 0.0526, | |
| "eval_samples_per_second": 1349.873, | |
| "eval_steps_per_second": 19.012, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.8825590014457703, | |
| "learning_rate": 0.0007, | |
| "loss": 0.746, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6906359791755676, | |
| "eval_runtime": 0.0518, | |
| "eval_samples_per_second": 1371.445, | |
| "eval_steps_per_second": 19.316, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.12335205078125, | |
| "learning_rate": 0.0006666666666666666, | |
| "loss": 0.6992, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6848593354225159, | |
| "eval_runtime": 0.0528, | |
| "eval_samples_per_second": 1345.421, | |
| "eval_steps_per_second": 18.95, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.29337310791015625, | |
| "learning_rate": 0.0006333333333333333, | |
| "loss": 0.6942, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.700924277305603, | |
| "eval_runtime": 0.0523, | |
| "eval_samples_per_second": 1358.042, | |
| "eval_steps_per_second": 19.127, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.3950302600860596, | |
| "learning_rate": 0.0006, | |
| "loss": 0.7, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.43661971830985913, | |
| "eval_loss": 0.6950923204421997, | |
| "eval_runtime": 0.0545, | |
| "eval_samples_per_second": 1303.189, | |
| "eval_steps_per_second": 18.355, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.581301748752594, | |
| "learning_rate": 0.0005666666666666667, | |
| "loss": 0.6976, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.685354471206665, | |
| "eval_runtime": 0.0529, | |
| "eval_samples_per_second": 1342.534, | |
| "eval_steps_per_second": 18.909, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.8408365845680237, | |
| "learning_rate": 0.0005333333333333334, | |
| "loss": 0.6999, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6901406645774841, | |
| "eval_runtime": 0.053, | |
| "eval_samples_per_second": 1340.63, | |
| "eval_steps_per_second": 18.882, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.37376636266708374, | |
| "learning_rate": 0.0005, | |
| "loss": 0.6948, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.6926164627075195, | |
| "eval_runtime": 0.0535, | |
| "eval_samples_per_second": 1327.152, | |
| "eval_steps_per_second": 18.692, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 45, | |
| "total_flos": 1253066401152000.0, | |
| "train_loss": 0.8229163063897027, | |
| "train_runtime": 48.1937, | |
| "train_samples_per_second": 395.28, | |
| "train_steps_per_second": 1.867 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1253066401152000.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |