| { | |
| "best_metric": 0.28307044506073, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-16000", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 16287, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09211495946941783, | |
| "grad_norm": 3.071866512298584, | |
| "learning_rate": 4.84647506755097e-07, | |
| "loss": 0.3448, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09211495946941783, | |
| "eval_loss": 1.0838947296142578, | |
| "eval_runtime": 76.5304, | |
| "eval_samples_per_second": 15.772, | |
| "eval_steps_per_second": 1.973, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "grad_norm": 2.2681832313537598, | |
| "learning_rate": 4.69295013510194e-07, | |
| "loss": 0.2708, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "eval_loss": 0.8948286175727844, | |
| "eval_runtime": 76.7259, | |
| "eval_samples_per_second": 15.731, | |
| "eval_steps_per_second": 1.968, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "grad_norm": 0.00790461990982294, | |
| "learning_rate": 4.5394252026529107e-07, | |
| "loss": 0.177, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "eval_loss": 0.8396429419517517, | |
| "eval_runtime": 78.5505, | |
| "eval_samples_per_second": 15.366, | |
| "eval_steps_per_second": 1.922, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "grad_norm": 0.29154184460639954, | |
| "learning_rate": 4.385900270203881e-07, | |
| "loss": 0.1831, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "eval_loss": 0.7720839977264404, | |
| "eval_runtime": 78.1799, | |
| "eval_samples_per_second": 15.439, | |
| "eval_steps_per_second": 1.931, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "grad_norm": 186.6424102783203, | |
| "learning_rate": 4.232375337754851e-07, | |
| "loss": 0.2038, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "eval_loss": 0.744644284248352, | |
| "eval_runtime": 76.5359, | |
| "eval_samples_per_second": 15.77, | |
| "eval_steps_per_second": 1.973, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "grad_norm": 0.009510820731520653, | |
| "learning_rate": 4.0788504053058217e-07, | |
| "loss": 0.1309, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "eval_loss": 0.7234784960746765, | |
| "eval_runtime": 76.6741, | |
| "eval_samples_per_second": 15.742, | |
| "eval_steps_per_second": 1.969, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "grad_norm": 542.5567626953125, | |
| "learning_rate": 3.925325472856792e-07, | |
| "loss": 0.1431, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "eval_loss": 0.6690872311592102, | |
| "eval_runtime": 76.4873, | |
| "eval_samples_per_second": 15.78, | |
| "eval_steps_per_second": 1.974, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "grad_norm": 3.6845791339874268, | |
| "learning_rate": 3.771800540407762e-07, | |
| "loss": 0.1411, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "eval_loss": 0.643679141998291, | |
| "eval_runtime": 76.4578, | |
| "eval_samples_per_second": 15.786, | |
| "eval_steps_per_second": 1.975, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "grad_norm": 0.005153482314199209, | |
| "learning_rate": 3.618275607958732e-07, | |
| "loss": 0.0849, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "eval_loss": 0.48050031065940857, | |
| "eval_runtime": 75.8694, | |
| "eval_samples_per_second": 15.909, | |
| "eval_steps_per_second": 1.99, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "grad_norm": 2.631653070449829, | |
| "learning_rate": 3.464750675509703e-07, | |
| "loss": 0.1026, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "eval_loss": 0.4800306260585785, | |
| "eval_runtime": 75.365, | |
| "eval_samples_per_second": 16.015, | |
| "eval_steps_per_second": 2.004, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "grad_norm": 1.8651682694326155e-05, | |
| "learning_rate": 3.311225743060673e-07, | |
| "loss": 0.1201, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "eval_loss": 0.4810214042663574, | |
| "eval_runtime": 74.984, | |
| "eval_samples_per_second": 16.097, | |
| "eval_steps_per_second": 2.014, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "grad_norm": 0.5627817511558533, | |
| "learning_rate": 3.157700810611643e-07, | |
| "loss": 0.0757, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "eval_loss": 0.46917983889579773, | |
| "eval_runtime": 74.883, | |
| "eval_samples_per_second": 16.118, | |
| "eval_steps_per_second": 2.016, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "grad_norm": 0.0001344973425148055, | |
| "learning_rate": 3.0041758781626137e-07, | |
| "loss": 0.0696, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "eval_loss": 0.47636929154396057, | |
| "eval_runtime": 76.4445, | |
| "eval_samples_per_second": 15.789, | |
| "eval_steps_per_second": 1.975, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "grad_norm": 0.007305172737687826, | |
| "learning_rate": 2.8506509457135833e-07, | |
| "loss": 0.0911, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "eval_loss": 0.4601423740386963, | |
| "eval_runtime": 76.451, | |
| "eval_samples_per_second": 15.788, | |
| "eval_steps_per_second": 1.975, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "grad_norm": 9.710428905407298e-09, | |
| "learning_rate": 2.697126013264554e-07, | |
| "loss": 0.0806, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "eval_loss": 0.45901384949684143, | |
| "eval_runtime": 74.7174, | |
| "eval_samples_per_second": 16.154, | |
| "eval_steps_per_second": 2.021, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "grad_norm": 0.008983594365417957, | |
| "learning_rate": 2.5436010808155247e-07, | |
| "loss": 0.088, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "eval_loss": 0.46537238359451294, | |
| "eval_runtime": 74.5146, | |
| "eval_samples_per_second": 16.198, | |
| "eval_steps_per_second": 2.026, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "grad_norm": 1.3851042240276001e-05, | |
| "learning_rate": 2.390076148366495e-07, | |
| "loss": 0.0878, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "eval_loss": 0.4768938720226288, | |
| "eval_runtime": 74.5757, | |
| "eval_samples_per_second": 16.185, | |
| "eval_steps_per_second": 2.025, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "grad_norm": 4.7310441004810855e-06, | |
| "learning_rate": 2.236551215917465e-07, | |
| "loss": 0.0369, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "eval_loss": 0.4683688282966614, | |
| "eval_runtime": 74.5774, | |
| "eval_samples_per_second": 16.185, | |
| "eval_steps_per_second": 2.025, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "grad_norm": 0.005283420439809561, | |
| "learning_rate": 2.083026283468435e-07, | |
| "loss": 0.1034, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "eval_loss": 0.4715610444545746, | |
| "eval_runtime": 74.6527, | |
| "eval_samples_per_second": 16.168, | |
| "eval_steps_per_second": 2.023, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "grad_norm": 1.703993320465088, | |
| "learning_rate": 1.9295013510194055e-07, | |
| "loss": 0.0852, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "eval_loss": 0.47203630208969116, | |
| "eval_runtime": 74.6773, | |
| "eval_samples_per_second": 16.163, | |
| "eval_steps_per_second": 2.022, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "grad_norm": 8.647094205116446e-07, | |
| "learning_rate": 1.7759764185703757e-07, | |
| "loss": 0.0493, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "eval_loss": 0.47136709094047546, | |
| "eval_runtime": 74.7402, | |
| "eval_samples_per_second": 16.149, | |
| "eval_steps_per_second": 2.02, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "grad_norm": 0.0002839279477484524, | |
| "learning_rate": 1.6224514861213458e-07, | |
| "loss": 0.0603, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "eval_loss": 0.4660906493663788, | |
| "eval_runtime": 74.8543, | |
| "eval_samples_per_second": 16.125, | |
| "eval_steps_per_second": 2.017, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "grad_norm": 0.17852088809013367, | |
| "learning_rate": 1.4689265536723165e-07, | |
| "loss": 0.0547, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "eval_loss": 0.4668720066547394, | |
| "eval_runtime": 74.8609, | |
| "eval_samples_per_second": 16.123, | |
| "eval_steps_per_second": 2.017, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "grad_norm": 4.607869868777925e-06, | |
| "learning_rate": 1.3154016212232866e-07, | |
| "loss": 0.0793, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "eval_loss": 0.4664075970649719, | |
| "eval_runtime": 74.7262, | |
| "eval_samples_per_second": 16.152, | |
| "eval_steps_per_second": 2.021, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.302449806594216, | |
| "grad_norm": 0.00044986297143623233, | |
| "learning_rate": 1.1625836556763061e-07, | |
| "loss": 0.0415, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.302449806594216, | |
| "eval_loss": 0.2888467311859131, | |
| "eval_runtime": 78.2614, | |
| "eval_samples_per_second": 15.41, | |
| "eval_steps_per_second": 1.929, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.394547798857985, | |
| "grad_norm": 4.1875861001017256e-08, | |
| "learning_rate": 1.0090870019033584e-07, | |
| "loss": 0.0565, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.394547798857985, | |
| "eval_loss": 0.29098838567733765, | |
| "eval_runtime": 78.3993, | |
| "eval_samples_per_second": 15.383, | |
| "eval_steps_per_second": 1.926, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4866457911217537, | |
| "grad_norm": 1.556837378302589e-05, | |
| "learning_rate": 8.555903481304106e-08, | |
| "loss": 0.0629, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.4866457911217537, | |
| "eval_loss": 0.2888970673084259, | |
| "eval_runtime": 76.7751, | |
| "eval_samples_per_second": 15.708, | |
| "eval_steps_per_second": 1.967, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.578743783385522, | |
| "grad_norm": 0.12299621850252151, | |
| "learning_rate": 7.020936943574628e-08, | |
| "loss": 0.0584, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.578743783385522, | |
| "eval_loss": 0.28743651509284973, | |
| "eval_runtime": 76.8684, | |
| "eval_samples_per_second": 15.689, | |
| "eval_steps_per_second": 1.964, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.670841775649291, | |
| "grad_norm": 0.25907593965530396, | |
| "learning_rate": 5.485970405845152e-08, | |
| "loss": 0.0582, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.670841775649291, | |
| "eval_loss": 0.2863319218158722, | |
| "eval_runtime": 77.0162, | |
| "eval_samples_per_second": 15.659, | |
| "eval_steps_per_second": 1.961, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.7629397679130596, | |
| "grad_norm": 1.9590417650761083e-05, | |
| "learning_rate": 3.951003868115675e-08, | |
| "loss": 0.052, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.7629397679130596, | |
| "eval_loss": 0.2845611274242401, | |
| "eval_runtime": 76.9043, | |
| "eval_samples_per_second": 15.682, | |
| "eval_steps_per_second": 1.963, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.8550377601768284, | |
| "grad_norm": 0.2627514600753784, | |
| "learning_rate": 2.4160373303861975e-08, | |
| "loss": 0.0402, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.8550377601768284, | |
| "eval_loss": 0.28350818157196045, | |
| "eval_runtime": 76.8887, | |
| "eval_samples_per_second": 15.685, | |
| "eval_steps_per_second": 1.964, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.947135752440597, | |
| "grad_norm": 0.00015808168973308057, | |
| "learning_rate": 8.8107079265672e-09, | |
| "loss": 0.0518, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.947135752440597, | |
| "eval_loss": 0.28307044506073, | |
| "eval_runtime": 76.733, | |
| "eval_samples_per_second": 15.717, | |
| "eval_steps_per_second": 1.968, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 16287, | |
| "total_flos": 5859923904498510.0, | |
| "train_loss": 0.014361034947702911, | |
| "train_runtime": 3221.5046, | |
| "train_samples_per_second": 10.11, | |
| "train_steps_per_second": 5.056 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16287, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 5859923904498510.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |