| { | |
| "best_metric": 0.09009132534265518, | |
| "best_model_checkpoint": "/kaggle/working/xls-r-amharic/checkpoint-2500", | |
| "epoch": 14.969696969696969, | |
| "eval_steps": 500, | |
| "global_step": 3705, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.40404040404040403, | |
| "grad_norm": 3.749224901199341, | |
| "learning_rate": 9.730094466936572e-06, | |
| "loss": 1.1928, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8080808080808081, | |
| "grad_norm": 5.272401809692383, | |
| "learning_rate": 9.46288798920378e-06, | |
| "loss": 0.7006, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "grad_norm": 2.211305618286133, | |
| "learning_rate": 9.192982456140351e-06, | |
| "loss": 0.3993, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6161616161616161, | |
| "grad_norm": 0.4073326289653778, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.2935, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0202020202020203, | |
| "grad_norm": 10.367269515991211, | |
| "learning_rate": 8.65587044534413e-06, | |
| "loss": 0.2847, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0202020202020203, | |
| "eval_accuracy": 0.9212121367454529, | |
| "eval_loss": 0.247890442609787, | |
| "eval_runtime": 43.8986, | |
| "eval_samples_per_second": 11.276, | |
| "eval_steps_per_second": 2.825, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4242424242424243, | |
| "grad_norm": 14.821730613708496, | |
| "learning_rate": 8.385964912280704e-06, | |
| "loss": 0.1788, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8282828282828283, | |
| "grad_norm": 1.658502221107483, | |
| "learning_rate": 8.116059379217275e-06, | |
| "loss": 0.1541, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.2323232323232323, | |
| "grad_norm": 7.924429416656494, | |
| "learning_rate": 7.846153846153847e-06, | |
| "loss": 0.1683, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.14956633746623993, | |
| "learning_rate": 7.576248313090419e-06, | |
| "loss": 0.1315, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.040404040404041, | |
| "grad_norm": 0.08813250064849854, | |
| "learning_rate": 7.306342780026991e-06, | |
| "loss": 0.1138, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.040404040404041, | |
| "eval_accuracy": 0.9434343576431274, | |
| "eval_loss": 0.20633606612682343, | |
| "eval_runtime": 43.6554, | |
| "eval_samples_per_second": 11.339, | |
| "eval_steps_per_second": 2.84, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 2.7388453483581543, | |
| "learning_rate": 7.036437246963563e-06, | |
| "loss": 0.1113, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.848484848484849, | |
| "grad_norm": 1.112307071685791, | |
| "learning_rate": 6.766531713900135e-06, | |
| "loss": 0.1174, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.252525252525253, | |
| "grad_norm": 0.9322103261947632, | |
| "learning_rate": 6.496626180836708e-06, | |
| "loss": 0.1025, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.656565656565657, | |
| "grad_norm": 0.03639671951532364, | |
| "learning_rate": 6.22672064777328e-06, | |
| "loss": 0.0754, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.0606060606060606, | |
| "grad_norm": 0.024523159489035606, | |
| "learning_rate": 5.956815114709852e-06, | |
| "loss": 0.0614, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0606060606060606, | |
| "eval_accuracy": 0.965656578540802, | |
| "eval_loss": 0.1415119469165802, | |
| "eval_runtime": 43.7636, | |
| "eval_samples_per_second": 11.311, | |
| "eval_steps_per_second": 2.833, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.4646464646464645, | |
| "grad_norm": 0.04833903908729553, | |
| "learning_rate": 5.686909581646424e-06, | |
| "loss": 0.0296, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.8686868686868685, | |
| "grad_norm": 0.018174033612012863, | |
| "learning_rate": 5.417004048582997e-06, | |
| "loss": 0.086, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.2865201532840729, | |
| "learning_rate": 5.147098515519568e-06, | |
| "loss": 0.0671, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.6767676767676765, | |
| "grad_norm": 0.0348142571747303, | |
| "learning_rate": 4.877192982456141e-06, | |
| "loss": 0.0512, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.080808080808081, | |
| "grad_norm": 0.2693362832069397, | |
| "learning_rate": 4.607287449392713e-06, | |
| "loss": 0.0349, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.080808080808081, | |
| "eval_accuracy": 0.973737359046936, | |
| "eval_loss": 0.13826610147953033, | |
| "eval_runtime": 43.9764, | |
| "eval_samples_per_second": 11.256, | |
| "eval_steps_per_second": 2.82, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.484848484848484, | |
| "grad_norm": 0.033031389117240906, | |
| "learning_rate": 4.337381916329285e-06, | |
| "loss": 0.0251, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 0.13385087251663208, | |
| "learning_rate": 4.067476383265857e-06, | |
| "loss": 0.0367, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 9.292929292929292, | |
| "grad_norm": 41.249717712402344, | |
| "learning_rate": 3.7975708502024296e-06, | |
| "loss": 0.0501, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 9.696969696969697, | |
| "grad_norm": 0.02456289902329445, | |
| "learning_rate": 3.527665317139002e-06, | |
| "loss": 0.0555, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 10.1010101010101, | |
| "grad_norm": 0.010708549991250038, | |
| "learning_rate": 3.2577597840755737e-06, | |
| "loss": 0.0143, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.1010101010101, | |
| "eval_accuracy": 0.9818181991577148, | |
| "eval_loss": 0.09009132534265518, | |
| "eval_runtime": 44.1262, | |
| "eval_samples_per_second": 11.218, | |
| "eval_steps_per_second": 2.81, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.505050505050505, | |
| "grad_norm": 0.012150867842137814, | |
| "learning_rate": 2.9905533063427807e-06, | |
| "loss": 0.0486, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.008297057822346687, | |
| "learning_rate": 2.7206477732793525e-06, | |
| "loss": 0.0349, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 11.313131313131313, | |
| "grad_norm": 0.013805734924972057, | |
| "learning_rate": 2.4507422402159244e-06, | |
| "loss": 0.0214, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 11.717171717171716, | |
| "grad_norm": 0.011365755461156368, | |
| "learning_rate": 2.180836707152497e-06, | |
| "loss": 0.0229, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 12.121212121212121, | |
| "grad_norm": 0.02992076426744461, | |
| "learning_rate": 1.910931174089069e-06, | |
| "loss": 0.0178, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.121212121212121, | |
| "eval_accuracy": 0.9777777791023254, | |
| "eval_loss": 0.1187622994184494, | |
| "eval_runtime": 43.9086, | |
| "eval_samples_per_second": 11.273, | |
| "eval_steps_per_second": 2.824, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 12.525252525252526, | |
| "grad_norm": 0.09338176250457764, | |
| "learning_rate": 1.6410256410256412e-06, | |
| "loss": 0.0252, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 12.929292929292929, | |
| "grad_norm": 0.27980300784111023, | |
| "learning_rate": 1.3738191632928477e-06, | |
| "loss": 0.0387, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.007048506755381823, | |
| "learning_rate": 1.1039136302294197e-06, | |
| "loss": 0.0093, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 13.737373737373737, | |
| "grad_norm": 0.00685643358156085, | |
| "learning_rate": 8.34008097165992e-07, | |
| "loss": 0.0222, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 14.141414141414142, | |
| "grad_norm": 0.14381413161754608, | |
| "learning_rate": 5.641025641025642e-07, | |
| "loss": 0.0222, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.141414141414142, | |
| "eval_accuracy": 0.9777777791023254, | |
| "eval_loss": 0.12370182573795319, | |
| "eval_runtime": 44.0309, | |
| "eval_samples_per_second": 11.242, | |
| "eval_steps_per_second": 2.816, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.015089770779013634, | |
| "learning_rate": 2.941970310391363e-07, | |
| "loss": 0.0304, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 14.94949494949495, | |
| "grad_norm": 1.67782723903656, | |
| "learning_rate": 2.4291497975708507e-08, | |
| "loss": 0.0173, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 14.969696969696969, | |
| "step": 3705, | |
| "total_flos": 3.163398064220592e+18, | |
| "train_loss": 0.13106194541521884, | |
| "train_runtime": 5199.6798, | |
| "train_samples_per_second": 5.709, | |
| "train_steps_per_second": 0.713 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 3705, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.163398064220592e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |