| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0129032258064514, | |
| "eval_steps": 10, | |
| "global_step": 116, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025806451612903226, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.89, | |
| "eval_samples_per_second": 73.037, | |
| "eval_steps_per_second": 10.113, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07741935483870968, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.15483870967741936, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23225806451612904, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.25806451612903225, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4804, | |
| "eval_samples_per_second": 135.309, | |
| "eval_steps_per_second": 18.735, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3096774193548387, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.991218658821608e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.3870967741935484, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.94520093661082e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4645161290322581, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.860114570402054e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5161290322580645, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4793, | |
| "eval_samples_per_second": 135.617, | |
| "eval_steps_per_second": 18.778, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5419354838709678, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.736631769270957e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6193548387096774, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.575728086215092e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6967741935483871, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.378674710978185e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7741935483870968, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.14702842725101e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7741935483870968, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4793, | |
| "eval_samples_per_second": 135.604, | |
| "eval_steps_per_second": 18.776, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8516129032258064, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.882619313590212e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.9290322580645162, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.587536285221656e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.0129032258064516, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.264110590952609e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.038709677419355, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4747, | |
| "eval_samples_per_second": 136.925, | |
| "eval_steps_per_second": 18.959, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0903225806451613, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.91489739557236e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.167741935483871, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.542655593246103e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.2451612903225806, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.150326011382604e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.2967741935483872, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4769, | |
| "eval_samples_per_second": 136.301, | |
| "eval_steps_per_second": 18.872, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3225806451612903, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.741008177171995e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.317935830345338e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.4774193548387098, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.884451375612865e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.5548387096774192, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4439794766146746e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5548387096774192, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4836, | |
| "eval_samples_per_second": 134.417, | |
| "eval_steps_per_second": 18.612, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.632258064516129, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.7096774193548387, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.5560205233853266e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.7870967741935484, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.115548624387137e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.8129032258064517, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4769, | |
| "eval_samples_per_second": 136.31, | |
| "eval_steps_per_second": 18.874, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.864516129032258, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.682064169654663e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.9419354838709677, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.258991822828007e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.0258064516129033, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.8496739886173995e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.07741935483871, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4796, | |
| "eval_samples_per_second": 135.533, | |
| "eval_steps_per_second": 18.766, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.1032258064516127, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4573444067538986e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.1806451612903226, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.0851026044276406e-05, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.258064516129032, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.7358894090473925e-05, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.335483870967742, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4124637147783432e-05, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.335483870967742, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4778, | |
| "eval_samples_per_second": 136.038, | |
| "eval_steps_per_second": 18.836, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.412903225806452, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1173806864097886e-05, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.490322580645161, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.529715727489912e-06, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.567741935483871, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.213252890218163e-06, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.5935483870967744, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4814, | |
| "eval_samples_per_second": 135.034, | |
| "eval_steps_per_second": 18.697, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6451612903225805, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.242719137849077e-06, | |
| "loss": 0.0, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.7225806451612904, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.6336823072904304e-06, | |
| "loss": 0.0, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.3988542959794627e-06, | |
| "loss": 0.0, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.8516129032258064, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.4762, | |
| "eval_samples_per_second": 136.507, | |
| "eval_steps_per_second": 18.901, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.8774193548387097, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.479906338917984e-07, | |
| "loss": 0.0, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.9548387096774196, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.781341178393244e-08, | |
| "loss": 0.0, | |
| "step": 114 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 447372853248.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |