| { | |
| "best_metric": 0.0367523655295372, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-390", | |
| "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": 4.539805889129639, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4713, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9548872180451128, | |
| "eval_loss": 0.09497568756341934, | |
| "eval_runtime": 1.0331, | |
| "eval_samples_per_second": 128.734, | |
| "eval_steps_per_second": 16.455, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.8214866518974304, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1717, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.08848864585161209, | |
| "eval_runtime": 0.9725, | |
| "eval_samples_per_second": 136.757, | |
| "eval_steps_per_second": 17.48, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 5.174630165100098, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1941, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.0367523655295372, | |
| "eval_runtime": 1.0358, | |
| "eval_samples_per_second": 128.405, | |
| "eval_steps_per_second": 16.413, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 6.90311336517334, | |
| "learning_rate": 0.0005, | |
| "loss": 0.178, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.0417233482003212, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1484, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.10054013133049011, | |
| "eval_runtime": 1.0366, | |
| "eval_samples_per_second": 128.308, | |
| "eval_steps_per_second": 16.4, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 3.222952365875244, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1574, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.06399582326412201, | |
| "eval_runtime": 1.1232, | |
| "eval_samples_per_second": 118.415, | |
| "eval_steps_per_second": 15.136, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 0.5326167345046997, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1948, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9172932330827067, | |
| "eval_loss": 0.18919076025485992, | |
| "eval_runtime": 1.1717, | |
| "eval_samples_per_second": 113.51, | |
| "eval_steps_per_second": 14.509, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 2.2912113666534424, | |
| "learning_rate": 0.0005, | |
| "loss": 0.2541, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 4.357283592224121, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4159, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.15302829444408417, | |
| "eval_runtime": 1.1496, | |
| "eval_samples_per_second": 115.692, | |
| "eval_steps_per_second": 14.788, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 2.716820240020752, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4951, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6842105263157895, | |
| "eval_loss": 0.7176167368888855, | |
| "eval_runtime": 0.9764, | |
| "eval_samples_per_second": 136.21, | |
| "eval_steps_per_second": 17.41, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 1.9368160963058472, | |
| "learning_rate": 0.0005, | |
| "loss": 0.6818, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 1.4557244777679443, | |
| "eval_runtime": 1.1951, | |
| "eval_samples_per_second": 111.285, | |
| "eval_steps_per_second": 14.224, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 1.1992454528808594, | |
| "learning_rate": 0.0005, | |
| "loss": 0.7561, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.8898727893829346, | |
| "learning_rate": 0.0005, | |
| "loss": 0.7749, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_loss": 0.7070097923278809, | |
| "eval_runtime": 1.1878, | |
| "eval_samples_per_second": 111.974, | |
| "eval_steps_per_second": 14.312, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.3764252471923828, | |
| "train_runtime": 205.4123, | |
| "train_samples_per_second": 50.338, | |
| "train_steps_per_second": 6.329 | |
| } | |
| ], | |
| "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 | |
| } | |