| { | |
| "best_metric": 0.3198860287666321, | |
| "best_model_checkpoint": "finetuned-fake-food/checkpoint-1800", | |
| "epoch": 2.5284450063211126, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1264222503160556, | |
| "grad_norm": 5.020864009857178, | |
| "learning_rate": 0.00019, | |
| "loss": 0.5416, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1264222503160556, | |
| "eval_accuracy": 0.7081468218442256, | |
| "eval_loss": 0.5592844486236572, | |
| "eval_runtime": 75.3793, | |
| "eval_samples_per_second": 14.818, | |
| "eval_steps_per_second": 1.857, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2528445006321112, | |
| "grad_norm": 3.241377115249634, | |
| "learning_rate": 0.00018, | |
| "loss": 0.5299, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2528445006321112, | |
| "eval_accuracy": 0.7421665174574754, | |
| "eval_loss": 0.5342246294021606, | |
| "eval_runtime": 75.6464, | |
| "eval_samples_per_second": 14.766, | |
| "eval_steps_per_second": 1.851, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.37926675094816686, | |
| "grad_norm": 3.0553033351898193, | |
| "learning_rate": 0.00017, | |
| "loss": 0.5503, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37926675094816686, | |
| "eval_accuracy": 0.7717099373321397, | |
| "eval_loss": 0.48751309514045715, | |
| "eval_runtime": 75.98, | |
| "eval_samples_per_second": 14.701, | |
| "eval_steps_per_second": 1.843, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5056890012642224, | |
| "grad_norm": 2.0104384422302246, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5561, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5056890012642224, | |
| "eval_accuracy": 0.7940913160250671, | |
| "eval_loss": 0.4621775448322296, | |
| "eval_runtime": 76.6055, | |
| "eval_samples_per_second": 14.581, | |
| "eval_steps_per_second": 1.828, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6321112515802781, | |
| "grad_norm": 3.4203433990478516, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.5581, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6321112515802781, | |
| "eval_accuracy": 0.7457475380483438, | |
| "eval_loss": 0.5501323342323303, | |
| "eval_runtime": 75.3664, | |
| "eval_samples_per_second": 14.821, | |
| "eval_steps_per_second": 1.858, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7585335018963337, | |
| "grad_norm": 1.4952611923217773, | |
| "learning_rate": 0.00014, | |
| "loss": 0.5845, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7585335018963337, | |
| "eval_accuracy": 0.747538048343778, | |
| "eval_loss": 0.5088097453117371, | |
| "eval_runtime": 76.5056, | |
| "eval_samples_per_second": 14.6, | |
| "eval_steps_per_second": 1.83, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8849557522123894, | |
| "grad_norm": 1.8074195384979248, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.5695, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8849557522123894, | |
| "eval_accuracy": 0.7860340196956133, | |
| "eval_loss": 0.4740249812602997, | |
| "eval_runtime": 76.9715, | |
| "eval_samples_per_second": 14.512, | |
| "eval_steps_per_second": 1.819, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.011378002528445, | |
| "grad_norm": 1.2785142660140991, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5406, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.011378002528445, | |
| "eval_accuracy": 0.7815577439570277, | |
| "eval_loss": 0.4855746030807495, | |
| "eval_runtime": 76.7685, | |
| "eval_samples_per_second": 14.55, | |
| "eval_steps_per_second": 1.824, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.1378002528445006, | |
| "grad_norm": 1.3373093605041504, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 0.5353, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.1378002528445006, | |
| "eval_accuracy": 0.8155774395702775, | |
| "eval_loss": 0.4251798987388611, | |
| "eval_runtime": 75.415, | |
| "eval_samples_per_second": 14.811, | |
| "eval_steps_per_second": 1.856, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2642225031605563, | |
| "grad_norm": 2.4060959815979004, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5345, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.2642225031605563, | |
| "eval_accuracy": 0.7761862130707251, | |
| "eval_loss": 0.50136399269104, | |
| "eval_runtime": 75.9241, | |
| "eval_samples_per_second": 14.712, | |
| "eval_steps_per_second": 1.844, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3906447534766118, | |
| "grad_norm": 1.6286314725875854, | |
| "learning_rate": 9e-05, | |
| "loss": 0.5105, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.3906447534766118, | |
| "eval_accuracy": 0.7860340196956133, | |
| "eval_loss": 0.48000478744506836, | |
| "eval_runtime": 75.3515, | |
| "eval_samples_per_second": 14.824, | |
| "eval_steps_per_second": 1.858, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.5170670037926675, | |
| "grad_norm": 2.462752103805542, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5266, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5170670037926675, | |
| "eval_accuracy": 0.7958818263205013, | |
| "eval_loss": 0.4617547392845154, | |
| "eval_runtime": 75.1188, | |
| "eval_samples_per_second": 14.87, | |
| "eval_steps_per_second": 1.864, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6434892541087232, | |
| "grad_norm": 2.6984634399414062, | |
| "learning_rate": 7e-05, | |
| "loss": 0.4709, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.6434892541087232, | |
| "eval_accuracy": 0.8281110116383169, | |
| "eval_loss": 0.39056020975112915, | |
| "eval_runtime": 74.801, | |
| "eval_samples_per_second": 14.933, | |
| "eval_steps_per_second": 1.872, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "grad_norm": 2.939568281173706, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4624, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.7699115044247788, | |
| "eval_accuracy": 0.8128916741271263, | |
| "eval_loss": 0.4208226203918457, | |
| "eval_runtime": 77.4109, | |
| "eval_samples_per_second": 14.429, | |
| "eval_steps_per_second": 1.809, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.8963337547408345, | |
| "grad_norm": 1.791272759437561, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4677, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.8963337547408345, | |
| "eval_accuracy": 0.8173679498657117, | |
| "eval_loss": 0.4207296073436737, | |
| "eval_runtime": 76.4178, | |
| "eval_samples_per_second": 14.617, | |
| "eval_steps_per_second": 1.832, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.02275600505689, | |
| "grad_norm": 1.7240327596664429, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4478, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.02275600505689, | |
| "eval_accuracy": 0.8478066248880931, | |
| "eval_loss": 0.35574597120285034, | |
| "eval_runtime": 75.4802, | |
| "eval_samples_per_second": 14.799, | |
| "eval_steps_per_second": 1.855, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.1491782553729455, | |
| "grad_norm": 3.029090642929077, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4451, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.1491782553729455, | |
| "eval_accuracy": 0.8442256042972247, | |
| "eval_loss": 0.3545984923839569, | |
| "eval_runtime": 75.7957, | |
| "eval_samples_per_second": 14.737, | |
| "eval_steps_per_second": 1.847, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.275600505689001, | |
| "grad_norm": 2.259213447570801, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3796, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.275600505689001, | |
| "eval_accuracy": 0.8719785138764548, | |
| "eval_loss": 0.3198860287666321, | |
| "eval_runtime": 74.5384, | |
| "eval_samples_per_second": 14.986, | |
| "eval_steps_per_second": 1.878, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.402022756005057, | |
| "grad_norm": 2.9328560829162598, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4358, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.402022756005057, | |
| "eval_accuracy": 0.8603401969561325, | |
| "eval_loss": 0.33084791898727417, | |
| "eval_runtime": 76.0815, | |
| "eval_samples_per_second": 14.682, | |
| "eval_steps_per_second": 1.84, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.5284450063211126, | |
| "grad_norm": 1.4755433797836304, | |
| "learning_rate": 0.0, | |
| "loss": 0.3373, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.5284450063211126, | |
| "eval_accuracy": 0.8540734109221128, | |
| "eval_loss": 0.34551766514778137, | |
| "eval_runtime": 75.3964, | |
| "eval_samples_per_second": 14.815, | |
| "eval_steps_per_second": 1.857, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.5284450063211126, | |
| "step": 2000, | |
| "total_flos": 1.2397168498542428e+18, | |
| "train_loss": 0.49920871353149415, | |
| "train_runtime": 3192.0436, | |
| "train_samples_per_second": 5.012, | |
| "train_steps_per_second": 0.627 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2397168498542428e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |