| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.256410256410255, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 485.6934509277344, | |
| "learning_rate": 9.9907e-06, | |
| "loss": 3.7897, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "eval_loss": 2.913217067718506, | |
| "eval_runtime": 36.1065, | |
| "eval_samples_per_second": 10.884, | |
| "eval_steps_per_second": 1.385, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 35.502872467041016, | |
| "learning_rate": 9.980700000000001e-06, | |
| "loss": 2.6449, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "eval_loss": 2.1936488151550293, | |
| "eval_runtime": 35.8618, | |
| "eval_samples_per_second": 10.959, | |
| "eval_steps_per_second": 1.394, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 44.5572624206543, | |
| "learning_rate": 9.970700000000001e-06, | |
| "loss": 2.123, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": 1.9975601434707642, | |
| "eval_runtime": 36.0033, | |
| "eval_samples_per_second": 10.916, | |
| "eval_steps_per_second": 1.389, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 91.06848907470703, | |
| "learning_rate": 9.960700000000001e-06, | |
| "loss": 1.9616, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "eval_loss": 1.9457911252975464, | |
| "eval_runtime": 35.8591, | |
| "eval_samples_per_second": 10.96, | |
| "eval_steps_per_second": 1.394, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 62.52458953857422, | |
| "learning_rate": 9.950700000000001e-06, | |
| "loss": 1.9329, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "eval_loss": 1.8983820676803589, | |
| "eval_runtime": 35.7607, | |
| "eval_samples_per_second": 10.99, | |
| "eval_steps_per_second": 1.398, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 46.58191680908203, | |
| "learning_rate": 9.940700000000001e-06, | |
| "loss": 1.8998, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": 1.8464109897613525, | |
| "eval_runtime": 35.5376, | |
| "eval_samples_per_second": 11.059, | |
| "eval_steps_per_second": 1.407, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.58974358974359, | |
| "grad_norm": 43.859432220458984, | |
| "learning_rate": 9.930700000000001e-06, | |
| "loss": 1.8632, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.58974358974359, | |
| "eval_loss": 1.8351906538009644, | |
| "eval_runtime": 35.8606, | |
| "eval_samples_per_second": 10.959, | |
| "eval_steps_per_second": 1.394, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "grad_norm": 36.12472152709961, | |
| "learning_rate": 9.920700000000001e-06, | |
| "loss": 1.8202, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "eval_loss": 1.7837625741958618, | |
| "eval_runtime": 35.8551, | |
| "eval_samples_per_second": 10.961, | |
| "eval_steps_per_second": 1.395, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 56.74811553955078, | |
| "learning_rate": 9.910700000000002e-06, | |
| "loss": 1.7767, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "eval_loss": 1.777076005935669, | |
| "eval_runtime": 35.652, | |
| "eval_samples_per_second": 11.023, | |
| "eval_steps_per_second": 1.402, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.128205128205128, | |
| "grad_norm": 97.7017822265625, | |
| "learning_rate": 9.900700000000002e-06, | |
| "loss": 1.7757, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.128205128205128, | |
| "eval_loss": 1.7516826391220093, | |
| "eval_runtime": 35.6867, | |
| "eval_samples_per_second": 11.013, | |
| "eval_steps_per_second": 1.401, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.641025641025641, | |
| "grad_norm": 21.942626953125, | |
| "learning_rate": 9.8907e-06, | |
| "loss": 1.7592, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.641025641025641, | |
| "eval_loss": 1.7459185123443604, | |
| "eval_runtime": 35.7882, | |
| "eval_samples_per_second": 10.981, | |
| "eval_steps_per_second": 1.397, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 31.696691513061523, | |
| "learning_rate": 9.8807e-06, | |
| "loss": 1.8098, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "eval_loss": 1.7705851793289185, | |
| "eval_runtime": 35.6241, | |
| "eval_samples_per_second": 11.032, | |
| "eval_steps_per_second": 1.404, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 69.51777648925781, | |
| "learning_rate": 9.8707e-06, | |
| "loss": 1.6903, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "eval_loss": 1.717505693435669, | |
| "eval_runtime": 35.7343, | |
| "eval_samples_per_second": 10.998, | |
| "eval_steps_per_second": 1.399, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.17948717948718, | |
| "grad_norm": 159.8116455078125, | |
| "learning_rate": 9.8607e-06, | |
| "loss": 1.7428, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.17948717948718, | |
| "eval_loss": 1.7039239406585693, | |
| "eval_runtime": 35.6477, | |
| "eval_samples_per_second": 11.025, | |
| "eval_steps_per_second": 1.403, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 52.08290100097656, | |
| "learning_rate": 9.8507e-06, | |
| "loss": 1.6913, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "eval_loss": 1.6931686401367188, | |
| "eval_runtime": 35.6627, | |
| "eval_samples_per_second": 11.02, | |
| "eval_steps_per_second": 1.402, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.205128205128204, | |
| "grad_norm": 67.04603576660156, | |
| "learning_rate": 9.8407e-06, | |
| "loss": 1.7072, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.205128205128204, | |
| "eval_loss": 1.6624616384506226, | |
| "eval_runtime": 35.7439, | |
| "eval_samples_per_second": 10.995, | |
| "eval_steps_per_second": 1.399, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.717948717948717, | |
| "grad_norm": 22.715839385986328, | |
| "learning_rate": 9.8307e-06, | |
| "loss": 1.6593, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.717948717948717, | |
| "eval_loss": 1.6604528427124023, | |
| "eval_runtime": 35.8353, | |
| "eval_samples_per_second": 10.967, | |
| "eval_steps_per_second": 1.395, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 36.652496337890625, | |
| "learning_rate": 9.8207e-06, | |
| "loss": 1.7015, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "eval_loss": 1.691542148590088, | |
| "eval_runtime": 35.7641, | |
| "eval_samples_per_second": 10.989, | |
| "eval_steps_per_second": 1.398, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.743589743589745, | |
| "grad_norm": 55.04361343383789, | |
| "learning_rate": 9.8107e-06, | |
| "loss": 1.6452, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 9.743589743589745, | |
| "eval_loss": 1.64271080493927, | |
| "eval_runtime": 35.6018, | |
| "eval_samples_per_second": 11.039, | |
| "eval_steps_per_second": 1.404, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.256410256410255, | |
| "grad_norm": 48.84671401977539, | |
| "learning_rate": 9.800700000000001e-06, | |
| "loss": 1.6642, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.256410256410255, | |
| "eval_loss": 1.6221907138824463, | |
| "eval_runtime": 35.7515, | |
| "eval_samples_per_second": 10.993, | |
| "eval_steps_per_second": 1.399, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 513, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.64486332416e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |