| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.30298439630359036, | |
| "eval_steps": 50, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00019, | |
| "loss": 1.4805, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 0.935647189617157, | |
| "eval_runtime": 1424.1427, | |
| "eval_samples_per_second": 9.27, | |
| "eval_steps_per_second": 1.159, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00018, | |
| "loss": 0.8948, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 0.8673275709152222, | |
| "eval_runtime": 1447.9268, | |
| "eval_samples_per_second": 9.118, | |
| "eval_steps_per_second": 1.14, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00017, | |
| "loss": 0.8357, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.8413857221603394, | |
| "eval_runtime": 1449.9965, | |
| "eval_samples_per_second": 9.105, | |
| "eval_steps_per_second": 1.139, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00016, | |
| "loss": 0.8461, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.8236712217330933, | |
| "eval_runtime": 1428.6867, | |
| "eval_samples_per_second": 9.241, | |
| "eval_steps_per_second": 1.156, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.8046, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.7915458679199219, | |
| "eval_runtime": 1422.1414, | |
| "eval_samples_per_second": 9.283, | |
| "eval_steps_per_second": 1.161, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00014, | |
| "loss": 0.744, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 0.7999407649040222, | |
| "eval_runtime": 1408.4646, | |
| "eval_samples_per_second": 9.373, | |
| "eval_steps_per_second": 1.172, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.7219, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 0.7075337171554565, | |
| "eval_runtime": 1407.3666, | |
| "eval_samples_per_second": 9.381, | |
| "eval_steps_per_second": 1.173, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00012, | |
| "loss": 0.7027, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.697420060634613, | |
| "eval_runtime": 1391.6821, | |
| "eval_samples_per_second": 9.486, | |
| "eval_steps_per_second": 1.186, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 0.6982, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 0.6917020678520203, | |
| "eval_runtime": 1383.9747, | |
| "eval_samples_per_second": 9.539, | |
| "eval_steps_per_second": 1.193, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6746, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.6870374083518982, | |
| "eval_runtime": 1381.4308, | |
| "eval_samples_per_second": 9.557, | |
| "eval_steps_per_second": 1.195, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9e-05, | |
| "loss": 0.6667, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.6836341619491577, | |
| "eval_runtime": 1384.702, | |
| "eval_samples_per_second": 9.534, | |
| "eval_steps_per_second": 1.192, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6872, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.6808720231056213, | |
| "eval_runtime": 1386.676, | |
| "eval_samples_per_second": 9.521, | |
| "eval_steps_per_second": 1.191, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 7e-05, | |
| "loss": 0.6793, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.6773844957351685, | |
| "eval_runtime": 1607.2067, | |
| "eval_samples_per_second": 8.214, | |
| "eval_steps_per_second": 1.027, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6989, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.6752211451530457, | |
| "eval_runtime": 1618.5926, | |
| "eval_samples_per_second": 8.156, | |
| "eval_steps_per_second": 1.02, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6525, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.6731519103050232, | |
| "eval_runtime": 1621.8237, | |
| "eval_samples_per_second": 8.14, | |
| "eval_steps_per_second": 1.018, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6722, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.6716203689575195, | |
| "eval_runtime": 1622.2693, | |
| "eval_samples_per_second": 8.138, | |
| "eval_steps_per_second": 1.018, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6687, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 0.6699801683425903, | |
| "eval_runtime": 1622.5334, | |
| "eval_samples_per_second": 8.137, | |
| "eval_steps_per_second": 1.018, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6381, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 0.6691889762878418, | |
| "eval_runtime": 1621.8507, | |
| "eval_samples_per_second": 8.14, | |
| "eval_steps_per_second": 1.018, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6632, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.6684426069259644, | |
| "eval_runtime": 1609.6365, | |
| "eval_samples_per_second": 8.202, | |
| "eval_steps_per_second": 1.026, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.0, | |
| "loss": 0.6683, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.668069064617157, | |
| "eval_runtime": 1595.5055, | |
| "eval_samples_per_second": 8.274, | |
| "eval_steps_per_second": 1.035, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1000, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "total_flos": 7.51250442533929e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |