| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 4.061954955173073e-06, | |
| "loss": 2.0922, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 5.28472523771611e-06, | |
| "loss": 1.5474, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "eval_loss": 2.08203125, | |
| "eval_runtime": 2.8721, | |
| "eval_samples_per_second": 3.83, | |
| "eval_steps_per_second": 0.696, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.8877, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.2622, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "eval_loss": 2.9765625, | |
| "eval_runtime": 2.8096, | |
| "eval_samples_per_second": 3.915, | |
| "eval_steps_per_second": 0.712, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0353, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0066, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.96484375, | |
| "eval_runtime": 2.8073, | |
| "eval_samples_per_second": 3.918, | |
| "eval_steps_per_second": 0.712, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0022, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.001, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "eval_loss": 4.140625, | |
| "eval_runtime": 2.8092, | |
| "eval_samples_per_second": 3.916, | |
| "eval_steps_per_second": 0.712, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0009, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0006, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 4.22265625, | |
| "eval_runtime": 2.8076, | |
| "eval_samples_per_second": 3.918, | |
| "eval_steps_per_second": 0.712, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 36.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0005, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0004, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 4.2890625, | |
| "eval_runtime": 2.8093, | |
| "eval_samples_per_second": 3.916, | |
| "eval_steps_per_second": 0.712, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 43.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 46.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 46.67, | |
| "eval_loss": 4.3515625, | |
| "eval_runtime": 2.8091, | |
| "eval_samples_per_second": 3.916, | |
| "eval_steps_per_second": 0.712, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 53.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 53.33, | |
| "eval_loss": 4.390625, | |
| "eval_runtime": 2.8088, | |
| "eval_samples_per_second": 3.916, | |
| "eval_steps_per_second": 0.712, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 56.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 4.42578125, | |
| "eval_runtime": 2.808, | |
| "eval_samples_per_second": 3.917, | |
| "eval_steps_per_second": 0.712, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 63.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "eval_loss": 4.453125, | |
| "eval_runtime": 2.8089, | |
| "eval_samples_per_second": 3.916, | |
| "eval_steps_per_second": 0.712, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 73.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 73.33, | |
| "eval_loss": 4.47265625, | |
| "eval_runtime": 2.8074, | |
| "eval_samples_per_second": 3.918, | |
| "eval_steps_per_second": 0.712, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 76.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 4.49609375, | |
| "eval_runtime": 2.8069, | |
| "eval_samples_per_second": 3.919, | |
| "eval_steps_per_second": 0.713, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 86.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 86.67, | |
| "eval_loss": 4.51953125, | |
| "eval_runtime": 2.8067, | |
| "eval_samples_per_second": 3.919, | |
| "eval_steps_per_second": 0.713, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 93.33, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 93.33, | |
| "eval_loss": 4.5390625, | |
| "eval_runtime": 2.8074, | |
| "eval_samples_per_second": 3.918, | |
| "eval_steps_per_second": 0.712, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 96.67, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0002, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0003, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 4.5546875, | |
| "eval_runtime": 2.8081, | |
| "eval_samples_per_second": 3.917, | |
| "eval_steps_per_second": 0.712, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 300, | |
| "total_flos": 28875518115840.0, | |
| "train_loss": 0.16138767729202907, | |
| "train_runtime": 5518.2957, | |
| "train_samples_per_second": 0.797, | |
| "train_steps_per_second": 0.054 | |
| } | |
| ], | |
| "max_steps": 300, | |
| "num_train_epochs": 100, | |
| "total_flos": 28875518115840.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |