| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.6680562793121418, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016675351745700884, | |
| "eval_loss": 11.763640403747559, | |
| "eval_runtime": 1.6992, | |
| "eval_samples_per_second": 119.468, | |
| "eval_steps_per_second": 60.028, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05002605523710266, | |
| "grad_norm": 0.028261149302124977, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 11.7625, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10005211047420531, | |
| "grad_norm": 0.028932636603713036, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 11.7641, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.15007816571130797, | |
| "grad_norm": 0.030681710690259933, | |
| "learning_rate": 6.75e-05, | |
| "loss": 11.7637, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.15007816571130797, | |
| "eval_loss": 11.76322078704834, | |
| "eval_runtime": 1.6869, | |
| "eval_samples_per_second": 120.339, | |
| "eval_steps_per_second": 60.466, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.20010422094841063, | |
| "grad_norm": 0.028992891311645508, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 11.7632, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2501302761855133, | |
| "grad_norm": 0.03157561272382736, | |
| "learning_rate": 0.0001125, | |
| "loss": 11.7637, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.30015633142261594, | |
| "grad_norm": 0.03268129751086235, | |
| "learning_rate": 0.000135, | |
| "loss": 11.7614, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.30015633142261594, | |
| "eval_loss": 11.761802673339844, | |
| "eval_runtime": 1.7013, | |
| "eval_samples_per_second": 119.322, | |
| "eval_steps_per_second": 59.955, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3501823866597186, | |
| "grad_norm": 0.03852235525846481, | |
| "learning_rate": 0.00014994217771805422, | |
| "loss": 11.7615, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.40020844189682125, | |
| "grad_norm": 0.03991653025150299, | |
| "learning_rate": 0.00014907662554463532, | |
| "loss": 11.7611, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.45023449713392394, | |
| "grad_norm": 0.044084977358579636, | |
| "learning_rate": 0.0001471841427340235, | |
| "loss": 11.7598, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.45023449713392394, | |
| "eval_loss": 11.759143829345703, | |
| "eval_runtime": 1.7109, | |
| "eval_samples_per_second": 118.649, | |
| "eval_steps_per_second": 59.617, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5002605523710266, | |
| "grad_norm": 0.05109262466430664, | |
| "learning_rate": 0.0001442909649383465, | |
| "loss": 11.7588, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5502866076081292, | |
| "grad_norm": 0.06546574831008911, | |
| "learning_rate": 0.0001404372005304598, | |
| "loss": 11.7578, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6003126628452319, | |
| "grad_norm": 0.06636377424001694, | |
| "learning_rate": 0.00013567627457812106, | |
| "loss": 11.7557, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6003126628452319, | |
| "eval_loss": 11.755020141601562, | |
| "eval_runtime": 1.7074, | |
| "eval_samples_per_second": 118.891, | |
| "eval_steps_per_second": 59.738, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6503387180823346, | |
| "grad_norm": 0.07919026911258698, | |
| "learning_rate": 0.0001300741882076764, | |
| "loss": 11.7547, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7003647733194373, | |
| "grad_norm": 0.10065369307994843, | |
| "learning_rate": 0.00012370860362476374, | |
| "loss": 11.751, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.7503908285565398, | |
| "grad_norm": 0.0932815745472908, | |
| "learning_rate": 0.00011666776747647015, | |
| "loss": 11.75, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7503908285565398, | |
| "eval_loss": 11.74897289276123, | |
| "eval_runtime": 1.7066, | |
| "eval_samples_per_second": 118.95, | |
| "eval_steps_per_second": 59.768, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8004168837936425, | |
| "grad_norm": 0.10969464480876923, | |
| "learning_rate": 0.00010904928748046599, | |
| "loss": 11.7494, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8504429390307452, | |
| "grad_norm": 0.09933030605316162, | |
| "learning_rate": 0.00010095877928081196, | |
| "loss": 11.7457, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.9004689942678479, | |
| "grad_norm": 0.10053879022598267, | |
| "learning_rate": 9.25084022891929e-05, | |
| "loss": 11.7445, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9004689942678479, | |
| "eval_loss": 11.742661476135254, | |
| "eval_runtime": 1.6933, | |
| "eval_samples_per_second": 119.883, | |
| "eval_steps_per_second": 60.237, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9504950495049505, | |
| "grad_norm": 0.08994650840759277, | |
| "learning_rate": 8.381530480933783e-05, | |
| "loss": 11.7425, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.0010422094841063, | |
| "grad_norm": 0.09454693645238876, | |
| "learning_rate": 7.5e-05, | |
| "loss": 12.021, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.051068264721209, | |
| "grad_norm": 0.09124528616666794, | |
| "learning_rate": 6.618469519066217e-05, | |
| "loss": 11.7571, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.051068264721209, | |
| "eval_loss": 11.738404273986816, | |
| "eval_runtime": 1.6952, | |
| "eval_samples_per_second": 119.753, | |
| "eval_steps_per_second": 60.171, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1010943199583116, | |
| "grad_norm": 0.08838286995887756, | |
| "learning_rate": 5.7491597710807114e-05, | |
| "loss": 11.7056, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.1511203751954142, | |
| "grad_norm": 0.08494899421930313, | |
| "learning_rate": 4.904122071918801e-05, | |
| "loss": 11.7037, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.201146430432517, | |
| "grad_norm": 0.07477749139070511, | |
| "learning_rate": 4.095071251953399e-05, | |
| "loss": 11.7745, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.201146430432517, | |
| "eval_loss": 11.73630428314209, | |
| "eval_runtime": 1.6967, | |
| "eval_samples_per_second": 119.645, | |
| "eval_steps_per_second": 60.117, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.2511724856696196, | |
| "grad_norm": 0.07697498053312302, | |
| "learning_rate": 3.333223252352985e-05, | |
| "loss": 11.7542, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.3011985409067224, | |
| "grad_norm": 0.08104263246059418, | |
| "learning_rate": 2.6291396375236232e-05, | |
| "loss": 11.7077, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.351224596143825, | |
| "grad_norm": 0.06006978452205658, | |
| "learning_rate": 1.99258117923236e-05, | |
| "loss": 11.8078, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.351224596143825, | |
| "eval_loss": 11.73535442352295, | |
| "eval_runtime": 1.6897, | |
| "eval_samples_per_second": 120.139, | |
| "eval_steps_per_second": 60.365, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.4012506513809275, | |
| "grad_norm": 0.053699202835559845, | |
| "learning_rate": 1.4323725421878949e-05, | |
| "loss": 11.5982, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.4512767066180303, | |
| "grad_norm": 0.07235908508300781, | |
| "learning_rate": 9.56279946954021e-06, | |
| "loss": 11.8316, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.501302761855133, | |
| "grad_norm": 0.05552632361650467, | |
| "learning_rate": 5.709035061653494e-06, | |
| "loss": 11.6771, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.501302761855133, | |
| "eval_loss": 11.735032081604004, | |
| "eval_runtime": 1.6936, | |
| "eval_samples_per_second": 119.863, | |
| "eval_steps_per_second": 60.227, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5513288170922355, | |
| "grad_norm": 0.07071811705827713, | |
| "learning_rate": 2.815857265976462e-06, | |
| "loss": 11.8002, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.6013548723293383, | |
| "grad_norm": 0.0655035451054573, | |
| "learning_rate": 9.233744553646754e-07, | |
| "loss": 11.716, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.6513809275664408, | |
| "grad_norm": 0.07924264669418335, | |
| "learning_rate": 5.7822281945782424e-08, | |
| "loss": 11.6701, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.6513809275664408, | |
| "eval_loss": 11.734976768493652, | |
| "eval_runtime": 1.6952, | |
| "eval_samples_per_second": 119.751, | |
| "eval_steps_per_second": 60.17, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 186318052392960.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |