| { | |
| "best_metric": 0.23085123300552368, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-130", | |
| "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": 1.3140615224838257, | |
| "learning_rate": 0.005, | |
| "loss": 0.4537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.23085123300552368, | |
| "eval_runtime": 0.9384, | |
| "eval_samples_per_second": 141.736, | |
| "eval_steps_per_second": 18.117, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.5738377571105957, | |
| "learning_rate": 0.005, | |
| "loss": 0.5034, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6165413533834586, | |
| "eval_loss": 1.0887765884399414, | |
| "eval_runtime": 0.9484, | |
| "eval_samples_per_second": 140.237, | |
| "eval_steps_per_second": 17.925, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 4.012147426605225, | |
| "learning_rate": 0.005, | |
| "loss": 0.506, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5714285714285714, | |
| "eval_loss": 0.9555807709693909, | |
| "eval_runtime": 0.9238, | |
| "eval_samples_per_second": 143.97, | |
| "eval_steps_per_second": 18.402, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.720919132232666, | |
| "learning_rate": 0.005, | |
| "loss": 0.9151, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.1517225205898285, | |
| "learning_rate": 0.005, | |
| "loss": 1.1528, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1074310541152954, | |
| "eval_runtime": 0.9649, | |
| "eval_samples_per_second": 137.839, | |
| "eval_steps_per_second": 17.618, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.09401340782642365, | |
| "learning_rate": 0.005, | |
| "loss": 1.1076, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.100753664970398, | |
| "eval_runtime": 0.982, | |
| "eval_samples_per_second": 135.436, | |
| "eval_steps_per_second": 17.311, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 0.5329942107200623, | |
| "learning_rate": 0.005, | |
| "loss": 1.1148, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1018885374069214, | |
| "eval_runtime": 0.9388, | |
| "eval_samples_per_second": 141.675, | |
| "eval_steps_per_second": 18.109, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 0.4093654453754425, | |
| "learning_rate": 0.005, | |
| "loss": 1.1095, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 0.4959578216075897, | |
| "learning_rate": 0.005, | |
| "loss": 1.1026, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1080141067504883, | |
| "eval_runtime": 1.0873, | |
| "eval_samples_per_second": 122.317, | |
| "eval_steps_per_second": 15.635, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 0.2822917699813843, | |
| "learning_rate": 0.005, | |
| "loss": 1.1073, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.0986037254333496, | |
| "eval_runtime": 0.9744, | |
| "eval_samples_per_second": 136.492, | |
| "eval_steps_per_second": 17.446, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 0.33672869205474854, | |
| "learning_rate": 0.005, | |
| "loss": 1.1009, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3383458646616541, | |
| "eval_loss": 1.0998892784118652, | |
| "eval_runtime": 0.9516, | |
| "eval_samples_per_second": 139.767, | |
| "eval_steps_per_second": 17.865, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.08797406405210495, | |
| "learning_rate": 0.005, | |
| "loss": 1.1118, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.42109444737434387, | |
| "learning_rate": 0.005, | |
| "loss": 1.1081, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.098858118057251, | |
| "eval_runtime": 0.9588, | |
| "eval_samples_per_second": 138.708, | |
| "eval_steps_per_second": 17.73, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.9533355126014122, | |
| "train_runtime": 174.5905, | |
| "train_samples_per_second": 59.224, | |
| "train_steps_per_second": 7.446 | |
| } | |
| ], | |
| "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": 8.772706474360013e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |