| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.042625745950554135, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0017050298380221654, | |
| "grad_norm": 1.6034964323043823, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4885, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0017050298380221654, | |
| "eval_loss": 1.3765087127685547, | |
| "eval_runtime": 115.0675, | |
| "eval_samples_per_second": 6.44, | |
| "eval_steps_per_second": 1.616, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0034100596760443308, | |
| "grad_norm": 1.3994951248168945, | |
| "learning_rate": 0.0001, | |
| "loss": 1.639, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005115089514066497, | |
| "grad_norm": 1.1658259630203247, | |
| "learning_rate": 9.565217391304348e-05, | |
| "loss": 1.2317, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005115089514066497, | |
| "eval_loss": 1.2974878549575806, | |
| "eval_runtime": 115.0701, | |
| "eval_samples_per_second": 6.44, | |
| "eval_steps_per_second": 1.616, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0068201193520886615, | |
| "grad_norm": 4.647302150726318, | |
| "learning_rate": 9.130434782608696e-05, | |
| "loss": 2.0665, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.008525149190110827, | |
| "grad_norm": 1.481662392616272, | |
| "learning_rate": 8.695652173913044e-05, | |
| "loss": 1.6932, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010230179028132993, | |
| "grad_norm": 3.079343795776367, | |
| "learning_rate": 8.260869565217392e-05, | |
| "loss": 1.7811, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010230179028132993, | |
| "eval_loss": 1.2517597675323486, | |
| "eval_runtime": 115.501, | |
| "eval_samples_per_second": 6.416, | |
| "eval_steps_per_second": 1.61, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011935208866155157, | |
| "grad_norm": 0.97172611951828, | |
| "learning_rate": 7.82608695652174e-05, | |
| "loss": 1.6069, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.013640238704177323, | |
| "grad_norm": 0.934044599533081, | |
| "learning_rate": 7.391304347826086e-05, | |
| "loss": 1.2855, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.015345268542199489, | |
| "grad_norm": 0.8091419339179993, | |
| "learning_rate": 6.956521739130436e-05, | |
| "loss": 1.1789, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.015345268542199489, | |
| "eval_loss": 1.2307988405227661, | |
| "eval_runtime": 115.0172, | |
| "eval_samples_per_second": 6.443, | |
| "eval_steps_per_second": 1.617, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.017050298380221655, | |
| "grad_norm": 0.9011499285697937, | |
| "learning_rate": 6.521739130434783e-05, | |
| "loss": 1.3867, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01875532821824382, | |
| "grad_norm": 0.7925646901130676, | |
| "learning_rate": 6.086956521739131e-05, | |
| "loss": 1.3662, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.020460358056265986, | |
| "grad_norm": 0.7750657796859741, | |
| "learning_rate": 5.652173913043478e-05, | |
| "loss": 1.2866, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.020460358056265986, | |
| "eval_loss": 1.2137305736541748, | |
| "eval_runtime": 115.3185, | |
| "eval_samples_per_second": 6.426, | |
| "eval_steps_per_second": 1.613, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02216538789428815, | |
| "grad_norm": 1.9830117225646973, | |
| "learning_rate": 5.217391304347826e-05, | |
| "loss": 1.6887, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.023870417732310314, | |
| "grad_norm": 0.6523770093917847, | |
| "learning_rate": 4.782608695652174e-05, | |
| "loss": 1.3654, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02557544757033248, | |
| "grad_norm": 0.7408842444419861, | |
| "learning_rate": 4.347826086956522e-05, | |
| "loss": 1.344, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02557544757033248, | |
| "eval_loss": 1.2061489820480347, | |
| "eval_runtime": 115.3493, | |
| "eval_samples_per_second": 6.424, | |
| "eval_steps_per_second": 1.612, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.027280477408354646, | |
| "grad_norm": 0.7093881964683533, | |
| "learning_rate": 3.91304347826087e-05, | |
| "loss": 1.3975, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.028985507246376812, | |
| "grad_norm": 0.8234415054321289, | |
| "learning_rate": 3.478260869565218e-05, | |
| "loss": 1.1462, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.030690537084398978, | |
| "grad_norm": 0.801357090473175, | |
| "learning_rate": 3.0434782608695656e-05, | |
| "loss": 1.4323, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.030690537084398978, | |
| "eval_loss": 1.2006877660751343, | |
| "eval_runtime": 115.0416, | |
| "eval_samples_per_second": 6.441, | |
| "eval_steps_per_second": 1.617, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03239556692242114, | |
| "grad_norm": 0.6534079313278198, | |
| "learning_rate": 2.608695652173913e-05, | |
| "loss": 1.3427, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03410059676044331, | |
| "grad_norm": 0.6739923357963562, | |
| "learning_rate": 2.173913043478261e-05, | |
| "loss": 1.3338, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03580562659846547, | |
| "grad_norm": 0.7460229396820068, | |
| "learning_rate": 1.739130434782609e-05, | |
| "loss": 1.2776, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03580562659846547, | |
| "eval_loss": 1.1980022192001343, | |
| "eval_runtime": 115.3914, | |
| "eval_samples_per_second": 6.422, | |
| "eval_steps_per_second": 1.612, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03751065643648764, | |
| "grad_norm": 0.8048145771026611, | |
| "learning_rate": 1.3043478260869566e-05, | |
| "loss": 1.3784, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0392156862745098, | |
| "grad_norm": 0.7263695001602173, | |
| "learning_rate": 8.695652173913044e-06, | |
| "loss": 1.5492, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.04092071611253197, | |
| "grad_norm": 0.693609356880188, | |
| "learning_rate": 4.347826086956522e-06, | |
| "loss": 1.2394, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04092071611253197, | |
| "eval_loss": 1.1968275308609009, | |
| "eval_runtime": 115.6679, | |
| "eval_samples_per_second": 6.406, | |
| "eval_steps_per_second": 1.608, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.042625745950554135, | |
| "grad_norm": 0.8267568945884705, | |
| "learning_rate": 0.0, | |
| "loss": 1.0747, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.6105266286166016e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |