| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.76923076923077, | |
| "eval_steps": 50, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "eval_loss": 10.376392364501953, | |
| "eval_runtime": 5.3202, | |
| "eval_samples_per_second": 282.131, | |
| "eval_steps_per_second": 4.511, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.09619140625, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 10.378, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.09716796875, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 10.3754, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.11279296875, | |
| "learning_rate": 0.0002, | |
| "loss": 10.3683, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.201171875, | |
| "learning_rate": 0.00019984815164333163, | |
| "loss": 10.3522, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 0.00019939306773179497, | |
| "loss": 10.3159, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_loss": 10.285228729248047, | |
| "eval_runtime": 5.2726, | |
| "eval_samples_per_second": 284.681, | |
| "eval_steps_per_second": 4.552, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 10.2518, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.00019757963826274357, | |
| "loss": 10.1828, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.00019622680003092503, | |
| "loss": 10.1177, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 10.0565, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.00019264940672148018, | |
| "loss": 9.998, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "eval_loss": 9.973793029785156, | |
| "eval_runtime": 5.2822, | |
| "eval_samples_per_second": 284.164, | |
| "eval_steps_per_second": 4.544, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.00019043571606975777, | |
| "loss": 9.942, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 9.8887, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 0.00018519194088383273, | |
| "loss": 9.836, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.0001821777815225245, | |
| "loss": 9.7845, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 9.7359, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "eval_loss": 9.719038963317871, | |
| "eval_runtime": 5.2736, | |
| "eval_samples_per_second": 284.626, | |
| "eval_steps_per_second": 4.551, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.00017541066097768963, | |
| "loss": 9.688, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 13.076923076923077, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 0.00017167825131684513, | |
| "loss": 9.6416, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 9.597, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 14.615384615384615, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.00016357237482099684, | |
| "loss": 9.5534, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 0.37890625, | |
| "learning_rate": 0.00015922352526649803, | |
| "loss": 9.5151, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "eval_loss": 9.504176139831543, | |
| "eval_runtime": 5.2799, | |
| "eval_samples_per_second": 284.283, | |
| "eval_steps_per_second": 4.545, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 16.153846153846153, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 9.4734, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "grad_norm": 0.388671875, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 9.4381, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 17.692307692307693, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 0.00014515333583108896, | |
| "loss": 9.4021, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 0.396484375, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 9.3725, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "grad_norm": 0.396484375, | |
| "learning_rate": 0.00013506375551927547, | |
| "loss": 9.3407, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "eval_loss": 9.341134071350098, | |
| "eval_runtime": 5.9586, | |
| "eval_samples_per_second": 251.904, | |
| "eval_steps_per_second": 4.028, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 0.00012985148110016947, | |
| "loss": 9.3153, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 20.76923076923077, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 9.2905, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 21.53846153846154, | |
| "grad_norm": 0.408203125, | |
| "learning_rate": 0.00011917106319237386, | |
| "loss": 9.2678, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 22.307692307692307, | |
| "grad_norm": 0.404296875, | |
| "learning_rate": 0.00011373535578184082, | |
| "loss": 9.2496, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 23.076923076923077, | |
| "grad_norm": 0.40625, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 9.2338, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 23.076923076923077, | |
| "eval_loss": 9.241479873657227, | |
| "eval_runtime": 5.2815, | |
| "eval_samples_per_second": 284.199, | |
| "eval_steps_per_second": 4.544, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 23.846153846153847, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 0.00010275543423681621, | |
| "loss": 9.2185, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 24.615384615384617, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 9.724456576318381e-05, | |
| "loss": 9.2101, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 25.384615384615383, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 9.2002, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 26.153846153846153, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 8.626464421815919e-05, | |
| "loss": 9.193, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 26.923076923076923, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 8.082893680762619e-05, | |
| "loss": 9.1896, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 26.923076923076923, | |
| "eval_loss": 9.20389461517334, | |
| "eval_runtime": 5.2643, | |
| "eval_samples_per_second": 285.129, | |
| "eval_steps_per_second": 4.559, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 27.692307692307693, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 9.186, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 28.46153846153846, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 7.014851889983057e-05, | |
| "loss": 9.182, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 29.23076923076923, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 6.493624448072457e-05, | |
| "loss": 9.1823, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 9.1806, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 5.484666416891109e-05, | |
| "loss": 9.18, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "eval_loss": 9.195965766906738, | |
| "eval_runtime": 5.2708, | |
| "eval_samples_per_second": 284.775, | |
| "eval_steps_per_second": 4.553, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 47, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 163663419801600.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |