| { | |
| "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": 5e-05, | |
| "loss": 1.1701, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1349, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 5e-05, | |
| "loss": 1.124, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0961, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0956, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 1.1443167924880981, | |
| "eval_runtime": 15.8615, | |
| "eval_samples_per_second": 16.518, | |
| "eval_steps_per_second": 4.161, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0875, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0766, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1265, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1126, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0635, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.124241828918457, | |
| "eval_runtime": 15.8746, | |
| "eval_samples_per_second": 16.504, | |
| "eval_steps_per_second": 4.158, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0434, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1071, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0926, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0677, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0208, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.1208317279815674, | |
| "eval_runtime": 15.9371, | |
| "eval_samples_per_second": 16.44, | |
| "eval_steps_per_second": 4.141, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9435, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0034, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9165, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8848, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9505, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.1453500986099243, | |
| "eval_runtime": 15.8776, | |
| "eval_samples_per_second": 16.501, | |
| "eval_steps_per_second": 4.157, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9259, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8383, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9317, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9336, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9604, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.1406688690185547, | |
| "eval_runtime": 15.8651, | |
| "eval_samples_per_second": 16.514, | |
| "eval_steps_per_second": 4.16, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8884, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9757, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8977, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8645, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 294, | |
| "total_flos": 7.119009482145792e+16, | |
| "train_loss": 1.0086434495692351, | |
| "train_runtime": 872.7075, | |
| "train_samples_per_second": 5.399, | |
| "train_steps_per_second": 0.337 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 250, | |
| "total_flos": 7.119009482145792e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |