| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.88888888888889, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 13.9056, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 14.53, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 3e-06, | |
| "loss": 14.6503, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 13.8701, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 5e-06, | |
| "loss": 12.7866, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 6e-06, | |
| "loss": 12.0329, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 11.6797, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 10.1652, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 9e-06, | |
| "loss": 8.5016, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1e-05, | |
| "loss": 7.3051, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 6.5266, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.2e-05, | |
| "loss": 4.9159, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 4.1691, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 2.9038, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.2194, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.97, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 1.4839, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.2807, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.0124, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6787, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.5985, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.463, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.4044, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.3232, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3017, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.2714, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.2657, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.1915, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.1732, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1623, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.1401, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.1644, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.1303, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.1083, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1249, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.1037, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.1048, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 10.56, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.0925, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.0976, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0796, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.0811, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0789, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.0788, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0674, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0632, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 12.78, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0779, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.0596, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.0568, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.0647, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0574, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 270683602944000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |