| { | |
| "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": 9.19404125213623, | |
| "learning_rate": 9.007246376811595e-06, | |
| "loss": 1.1881, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 5.967567567567568, | |
| "eval_accuracy": 0.08407437348423606, | |
| "eval_loss": 2.9810373783111572, | |
| "eval_runtime": 4.7675, | |
| "eval_samples_per_second": 259.466, | |
| "eval_steps_per_second": 13.005, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 11.935135135135136, | |
| "grad_norm": 19.058837890625, | |
| "learning_rate": 8.007246376811595e-06, | |
| "loss": 0.9134, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 11.935135135135136, | |
| "eval_accuracy": 0.09458367016976556, | |
| "eval_loss": 3.3983819484710693, | |
| "eval_runtime": 4.7598, | |
| "eval_samples_per_second": 259.887, | |
| "eval_steps_per_second": 13.026, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 17.9027027027027, | |
| "grad_norm": 16.894025802612305, | |
| "learning_rate": 7.007246376811595e-06, | |
| "loss": 0.6902, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 17.9027027027027, | |
| "eval_accuracy": 0.09458367016976556, | |
| "eval_loss": 3.673595905303955, | |
| "eval_runtime": 4.6701, | |
| "eval_samples_per_second": 264.877, | |
| "eval_steps_per_second": 13.276, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 23.87027027027027, | |
| "grad_norm": 13.978982925415039, | |
| "learning_rate": 6.007246376811595e-06, | |
| "loss": 0.6084, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 23.87027027027027, | |
| "eval_accuracy": 0.09417946645109135, | |
| "eval_loss": 3.7337164878845215, | |
| "eval_runtime": 4.758, | |
| "eval_samples_per_second": 259.982, | |
| "eval_steps_per_second": 13.031, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 29.83783783783784, | |
| "grad_norm": 4.979071617126465, | |
| "learning_rate": 5.0072463768115946e-06, | |
| "loss": 0.5615, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 29.83783783783784, | |
| "eval_accuracy": 0.09329021827000808, | |
| "eval_loss": 3.6272082328796387, | |
| "eval_runtime": 4.7242, | |
| "eval_samples_per_second": 261.842, | |
| "eval_steps_per_second": 13.124, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 35.8054054054054, | |
| "grad_norm": 7.468260288238525, | |
| "learning_rate": 4.007246376811594e-06, | |
| "loss": 0.5358, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 35.8054054054054, | |
| "eval_accuracy": 0.09431420102398276, | |
| "eval_loss": 3.650233507156372, | |
| "eval_runtime": 4.8229, | |
| "eval_samples_per_second": 256.486, | |
| "eval_steps_per_second": 12.855, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 41.77297297297297, | |
| "grad_norm": 6.302524566650391, | |
| "learning_rate": 3.0072463768115946e-06, | |
| "loss": 0.5184, | |
| "step": 1932 | |
| }, | |
| { | |
| "epoch": 41.77297297297297, | |
| "eval_accuracy": 0.09389074950918119, | |
| "eval_loss": 3.6430163383483887, | |
| "eval_runtime": 4.6822, | |
| "eval_samples_per_second": 264.19, | |
| "eval_steps_per_second": 13.242, | |
| "step": 1932 | |
| }, | |
| { | |
| "epoch": 47.74054054054054, | |
| "grad_norm": 3.3958232402801514, | |
| "learning_rate": 2.0072463768115945e-06, | |
| "loss": 0.5087, | |
| "step": 2208 | |
| }, | |
| { | |
| "epoch": 47.74054054054054, | |
| "eval_accuracy": 0.09458367016976556, | |
| "eval_loss": 3.6398158073425293, | |
| "eval_runtime": 4.7183, | |
| "eval_samples_per_second": 262.169, | |
| "eval_steps_per_second": 13.14, | |
| "step": 2208 | |
| }, | |
| { | |
| "epoch": 53.70810810810811, | |
| "grad_norm": 5.994683742523193, | |
| "learning_rate": 1.0108695652173914e-06, | |
| "loss": 0.4951, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 53.70810810810811, | |
| "eval_accuracy": 0.09485313931554837, | |
| "eval_loss": 3.7279467582702637, | |
| "eval_runtime": 4.7161, | |
| "eval_samples_per_second": 262.292, | |
| "eval_steps_per_second": 13.146, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "grad_norm": 3.0481512546539307, | |
| "learning_rate": 1.0869565217391305e-08, | |
| "loss": 0.4939, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "eval_accuracy": 0.0952303961196443, | |
| "eval_loss": 3.7438390254974365, | |
| "eval_runtime": 4.7152, | |
| "eval_samples_per_second": 262.343, | |
| "eval_steps_per_second": 13.149, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "step": 2760, | |
| "total_flos": 2.1787658809800192e+17, | |
| "train_loss": 0.6513540627299875, | |
| "train_runtime": 4684.2441, | |
| "train_samples_per_second": 118.252, | |
| "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": 2.1787658809800192e+17, | |
| "train_batch_size": 25, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |