| { | |
| "best_metric": 0.5416202545166016, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-1170", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 3.9101290702819824, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1012, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5112781954887218, | |
| "eval_loss": 1.0127754211425781, | |
| "eval_runtime": 1.0349, | |
| "eval_samples_per_second": 128.515, | |
| "eval_steps_per_second": 16.427, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.3854057788848877, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0265, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6090225563909775, | |
| "eval_loss": 0.8759572505950928, | |
| "eval_runtime": 1.0786, | |
| "eval_samples_per_second": 123.311, | |
| "eval_steps_per_second": 15.762, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 2.5720674991607666, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9263, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6390977443609023, | |
| "eval_loss": 0.8891911506652832, | |
| "eval_runtime": 1.1187, | |
| "eval_samples_per_second": 118.885, | |
| "eval_steps_per_second": 15.196, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 4.826686382293701, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8131, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 2.224466323852539, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7348, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.706766917293233, | |
| "eval_loss": 0.7197765111923218, | |
| "eval_runtime": 1.2623, | |
| "eval_samples_per_second": 105.362, | |
| "eval_steps_per_second": 13.467, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 3.2183878421783447, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6856, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.706766917293233, | |
| "eval_loss": 0.7358166575431824, | |
| "eval_runtime": 1.192, | |
| "eval_samples_per_second": 111.582, | |
| "eval_steps_per_second": 14.262, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 5.501096248626709, | |
| "learning_rate": 5e-05, | |
| "loss": 0.728, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_loss": 0.6856114268302917, | |
| "eval_runtime": 1.0117, | |
| "eval_samples_per_second": 131.468, | |
| "eval_steps_per_second": 16.804, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 3.860182523727417, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6488, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 7.866222858428955, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6221, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6917293233082706, | |
| "eval_loss": 0.8288138508796692, | |
| "eval_runtime": 1.7977, | |
| "eval_samples_per_second": 73.984, | |
| "eval_steps_per_second": 9.457, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 3.9533588886260986, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6347, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7368421052631579, | |
| "eval_loss": 0.5764672756195068, | |
| "eval_runtime": 1.0499, | |
| "eval_samples_per_second": 126.675, | |
| "eval_steps_per_second": 16.192, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 9.031426429748535, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6217, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7744360902255639, | |
| "eval_loss": 0.5416202545166016, | |
| "eval_runtime": 1.0499, | |
| "eval_samples_per_second": 126.682, | |
| "eval_steps_per_second": 16.192, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 7.125640392303467, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6099, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 15.404199600219727, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5615, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7894736842105263, | |
| "eval_loss": 0.5903372764587402, | |
| "eval_runtime": 1.0014, | |
| "eval_samples_per_second": 132.812, | |
| "eval_steps_per_second": 16.976, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.400578669044531e+17, | |
| "train_loss": 0.7472451694195087, | |
| "train_runtime": 219.7142, | |
| "train_samples_per_second": 47.061, | |
| "train_steps_per_second": 5.917 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 8.400578669044531e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |