| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.037404151860856556, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007480830372171312, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007480830372171312, | |
| "eval_loss": NaN, | |
| "eval_runtime": 13.6052, | |
| "eval_samples_per_second": 41.381, | |
| "eval_steps_per_second": 20.727, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014961660744342623, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0022442491116513932, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0029923321488685246, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0029923321488685246, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8253, | |
| "eval_samples_per_second": 43.897, | |
| "eval_steps_per_second": 21.988, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0037404151860856555, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0044884982233027865, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005236581260519918, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005984664297737049, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005984664297737049, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8349, | |
| "eval_samples_per_second": 43.865, | |
| "eval_steps_per_second": 21.971, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00673274733495418, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007480830372171311, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008228913409388442, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.008976996446605573, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008976996446605573, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8361, | |
| "eval_samples_per_second": 43.861, | |
| "eval_steps_per_second": 21.969, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.009725079483822705, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010473162521039836, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011221245558256966, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011969328595474098, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.011969328595474098, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8344, | |
| "eval_samples_per_second": 43.867, | |
| "eval_steps_per_second": 21.972, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.012717411632691229, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01346549466990836, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014213577707125492, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.014961660744342622, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.014961660744342622, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8332, | |
| "eval_samples_per_second": 43.871, | |
| "eval_steps_per_second": 21.974, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015709743781559753, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.016457826818776885, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.017205909855994014, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.017953992893211146, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.017953992893211146, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8444, | |
| "eval_samples_per_second": 43.832, | |
| "eval_steps_per_second": 21.955, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.018702075930428278, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01945015896764541, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.02019824200486254, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.02094632504207967, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.02094632504207967, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8418, | |
| "eval_samples_per_second": 43.841, | |
| "eval_steps_per_second": 21.96, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.021694408079296804, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.022442491116513932, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.023190574153731065, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.023938657190948197, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.023938657190948197, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8963, | |
| "eval_samples_per_second": 43.656, | |
| "eval_steps_per_second": 21.867, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.024686740228165326, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.025434823265382458, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.02618290630259959, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02693098933981672, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02693098933981672, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8548, | |
| "eval_samples_per_second": 43.797, | |
| "eval_steps_per_second": 21.937, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02767907237703385, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.028427155414250983, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.029175238451468112, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.029923321488685244, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.029923321488685244, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.8946, | |
| "eval_samples_per_second": 43.662, | |
| "eval_steps_per_second": 21.87, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.030671404525902377, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.031419487563119505, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.032167570600336634, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03291565363755377, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03291565363755377, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.879, | |
| "eval_samples_per_second": 43.714, | |
| "eval_steps_per_second": 21.896, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0336637366747709, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03441181971198803, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03515990274920516, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.03590798578642229, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03590798578642229, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.9052, | |
| "eval_samples_per_second": 43.626, | |
| "eval_steps_per_second": 21.852, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03665606882363942, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.037404151860856556, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2419306423910400.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |