| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.99660441426146, | |
| "eval_steps": 50, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2065, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1456, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1411, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1179, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1077, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 1.1505802869796753, | |
| "eval_runtime": 15.5075, | |
| "eval_samples_per_second": 16.895, | |
| "eval_steps_per_second": 4.256, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0964, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0923, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1404, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1145, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0642, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.1310365200042725, | |
| "eval_runtime": 15.5192, | |
| "eval_samples_per_second": 16.882, | |
| "eval_steps_per_second": 4.253, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0558, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1053, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0948, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0725, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0399, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.1266452074050903, | |
| "eval_runtime": 15.5722, | |
| "eval_samples_per_second": 16.825, | |
| "eval_steps_per_second": 4.238, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9873, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0614, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 3e-05, | |
| "loss": 0.979, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9585, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0165, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.1301801204681396, | |
| "eval_runtime": 15.5114, | |
| "eval_samples_per_second": 16.891, | |
| "eval_steps_per_second": 4.255, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 3e-05, | |
| "loss": 0.987, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9076, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9922, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9889, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0272, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.1314338445663452, | |
| "eval_runtime": 15.5119, | |
| "eval_samples_per_second": 16.89, | |
| "eval_steps_per_second": 4.255, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3e-05, | |
| "loss": 0.977, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 3e-05, | |
| "loss": 1.0425, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9611, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9305, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 294, | |
| "total_flos": 6.900509298543821e+16, | |
| "train_loss": 1.0460854338950851, | |
| "train_runtime": 864.371, | |
| "train_samples_per_second": 5.451, | |
| "train_steps_per_second": 0.34 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 250, | |
| "total_flos": 6.900509298543821e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |