| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.67567567567568, | |
| "eval_steps": 276, | |
| "global_step": 2760, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.967567567567568, | |
| "grad_norm": 6.747945308685303, | |
| "learning_rate": 9.007246376811595e-06, | |
| "loss": 1.168, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 5.967567567567568, | |
| "eval_accuracy": 0.10670978172999192, | |
| "eval_loss": 2.970684051513672, | |
| "eval_runtime": 4.6739, | |
| "eval_samples_per_second": 264.663, | |
| "eval_steps_per_second": 13.265, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 11.935135135135136, | |
| "grad_norm": 10.996785163879395, | |
| "learning_rate": 8.007246376811595e-06, | |
| "loss": 0.8242, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 11.935135135135136, | |
| "eval_accuracy": 0.09943411479385611, | |
| "eval_loss": 3.4871609210968018, | |
| "eval_runtime": 4.7689, | |
| "eval_samples_per_second": 259.391, | |
| "eval_steps_per_second": 13.001, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 17.9027027027027, | |
| "grad_norm": 11.926712036132812, | |
| "learning_rate": 7.007246376811595e-06, | |
| "loss": 0.639, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 17.9027027027027, | |
| "eval_accuracy": 0.09646995419024522, | |
| "eval_loss": 3.507148504257202, | |
| "eval_runtime": 4.791, | |
| "eval_samples_per_second": 258.19, | |
| "eval_steps_per_second": 12.941, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 23.87027027027027, | |
| "grad_norm": 6.665815830230713, | |
| "learning_rate": 6.007246376811595e-06, | |
| "loss": 0.5825, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 23.87027027027027, | |
| "eval_accuracy": 0.09943411479385611, | |
| "eval_loss": 3.5618715286254883, | |
| "eval_runtime": 4.7419, | |
| "eval_samples_per_second": 260.867, | |
| "eval_steps_per_second": 13.075, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 29.83783783783784, | |
| "grad_norm": 3.5852649211883545, | |
| "learning_rate": 5.0072463768115946e-06, | |
| "loss": 0.5449, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 29.83783783783784, | |
| "eval_accuracy": 0.09943411479385611, | |
| "eval_loss": 3.5003042221069336, | |
| "eval_runtime": 4.7658, | |
| "eval_samples_per_second": 259.56, | |
| "eval_steps_per_second": 13.009, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 35.8054054054054, | |
| "grad_norm": 4.666769981384277, | |
| "learning_rate": 4.007246376811594e-06, | |
| "loss": 0.5257, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 35.8054054054054, | |
| "eval_accuracy": 0.10185933710590138, | |
| "eval_loss": 3.478468656539917, | |
| "eval_runtime": 4.7509, | |
| "eval_samples_per_second": 260.373, | |
| "eval_steps_per_second": 13.05, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 41.77297297297297, | |
| "grad_norm": 3.8682823181152344, | |
| "learning_rate": 3.0072463768115946e-06, | |
| "loss": 0.512, | |
| "step": 1932 | |
| }, | |
| { | |
| "epoch": 41.77297297297297, | |
| "eval_accuracy": 0.10197482388266543, | |
| "eval_loss": 3.463022470474243, | |
| "eval_runtime": 4.7286, | |
| "eval_samples_per_second": 261.601, | |
| "eval_steps_per_second": 13.112, | |
| "step": 1932 | |
| }, | |
| { | |
| "epoch": 47.74054054054054, | |
| "grad_norm": 2.4376981258392334, | |
| "learning_rate": 2.0108695652173916e-06, | |
| "loss": 0.5039, | |
| "step": 2208 | |
| }, | |
| { | |
| "epoch": 47.74054054054054, | |
| "eval_accuracy": 0.10357720291026677, | |
| "eval_loss": 3.5130271911621094, | |
| "eval_runtime": 4.6912, | |
| "eval_samples_per_second": 263.685, | |
| "eval_steps_per_second": 13.216, | |
| "step": 2208 | |
| }, | |
| { | |
| "epoch": 53.70810810810811, | |
| "grad_norm": 3.7486321926116943, | |
| "learning_rate": 1.0108695652173914e-06, | |
| "loss": 0.492, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 53.70810810810811, | |
| "eval_accuracy": 0.1038354441749753, | |
| "eval_loss": 3.5763065814971924, | |
| "eval_runtime": 4.7064, | |
| "eval_samples_per_second": 262.834, | |
| "eval_steps_per_second": 13.174, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "grad_norm": 2.842862129211426, | |
| "learning_rate": 1.0869565217391305e-08, | |
| "loss": 0.4922, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "eval_accuracy": 0.10412287793047696, | |
| "eval_loss": 3.5836520195007324, | |
| "eval_runtime": 4.7742, | |
| "eval_samples_per_second": 259.102, | |
| "eval_steps_per_second": 12.987, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "step": 2760, | |
| "total_flos": 2.1765475115811533e+17, | |
| "train_loss": 0.628451527028844, | |
| "train_runtime": 4688.2404, | |
| "train_samples_per_second": 118.151, | |
| "train_steps_per_second": 0.589 | |
| } | |
| ], | |
| "logging_steps": 276, | |
| "max_steps": 2760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.1765475115811533e+17, | |
| "train_batch_size": 25, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |