| { | |
| "best_metric": 0.5308398008346558, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l57-l/checkpoint-16000", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 16284, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09211495946941783, | |
| "grad_norm": 2.6780805587768555, | |
| "learning_rate": 4.84647506755097e-07, | |
| "loss": 0.2705, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09211495946941783, | |
| "eval_loss": 1.0680681467056274, | |
| "eval_runtime": 74.5723, | |
| "eval_samples_per_second": 16.186, | |
| "eval_steps_per_second": 2.025, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "grad_norm": 8.301130294799805, | |
| "learning_rate": 4.69295013510194e-07, | |
| "loss": 0.2545, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18422991893883567, | |
| "eval_loss": 0.9443553686141968, | |
| "eval_runtime": 74.9643, | |
| "eval_samples_per_second": 16.101, | |
| "eval_steps_per_second": 2.014, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "grad_norm": 0.018898434937000275, | |
| "learning_rate": 4.5394252026529107e-07, | |
| "loss": 0.234, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2763448784082535, | |
| "eval_loss": 0.876915693283081, | |
| "eval_runtime": 75.6401, | |
| "eval_samples_per_second": 15.957, | |
| "eval_steps_per_second": 1.996, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "grad_norm": 0.00010984255641233176, | |
| "learning_rate": 4.385900270203881e-07, | |
| "loss": 0.1539, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.36845983787767134, | |
| "eval_loss": 0.8415330648422241, | |
| "eval_runtime": 76.9959, | |
| "eval_samples_per_second": 15.676, | |
| "eval_steps_per_second": 1.961, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "grad_norm": 395.890625, | |
| "learning_rate": 4.232375337754851e-07, | |
| "loss": 0.1766, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46057479734708917, | |
| "eval_loss": 0.7660450339317322, | |
| "eval_runtime": 76.733, | |
| "eval_samples_per_second": 15.73, | |
| "eval_steps_per_second": 1.968, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "grad_norm": 363.6924743652344, | |
| "learning_rate": 4.0788504053058217e-07, | |
| "loss": 0.1679, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.552689756816507, | |
| "eval_loss": 0.7269155383110046, | |
| "eval_runtime": 76.7928, | |
| "eval_samples_per_second": 15.718, | |
| "eval_steps_per_second": 1.966, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "grad_norm": 28.914026260375977, | |
| "learning_rate": 3.925325472856792e-07, | |
| "loss": 0.1104, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6448047162859248, | |
| "eval_loss": 0.7097663283348083, | |
| "eval_runtime": 76.6897, | |
| "eval_samples_per_second": 15.739, | |
| "eval_steps_per_second": 1.969, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "grad_norm": 116.85651397705078, | |
| "learning_rate": 3.771800540407762e-07, | |
| "loss": 0.1367, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7369196757553427, | |
| "eval_loss": 0.6968969106674194, | |
| "eval_runtime": 77.1872, | |
| "eval_samples_per_second": 15.637, | |
| "eval_steps_per_second": 1.956, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "grad_norm": 391.0343322753906, | |
| "learning_rate": 3.618275607958732e-07, | |
| "loss": 0.1129, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8290346352247605, | |
| "eval_loss": 0.6776800751686096, | |
| "eval_runtime": 75.0351, | |
| "eval_samples_per_second": 16.086, | |
| "eval_steps_per_second": 2.012, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "grad_norm": 2.0514346033451147e-05, | |
| "learning_rate": 3.464750675509703e-07, | |
| "loss": 0.1125, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9211495946941783, | |
| "eval_loss": 0.6658498048782349, | |
| "eval_runtime": 77.111, | |
| "eval_samples_per_second": 15.653, | |
| "eval_steps_per_second": 1.958, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "grad_norm": 0.000568103976547718, | |
| "learning_rate": 3.311225743060673e-07, | |
| "loss": 0.1071, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013264554163596, | |
| "eval_loss": 0.6465296149253845, | |
| "eval_runtime": 75.2951, | |
| "eval_samples_per_second": 16.03, | |
| "eval_steps_per_second": 2.005, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "grad_norm": 0.02698369137942791, | |
| "learning_rate": 3.157700810611643e-07, | |
| "loss": 0.0553, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.105379513633014, | |
| "eval_loss": 0.6357071995735168, | |
| "eval_runtime": 76.6726, | |
| "eval_samples_per_second": 15.742, | |
| "eval_steps_per_second": 1.969, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "grad_norm": 0.0005627564387395978, | |
| "learning_rate": 3.0041758781626137e-07, | |
| "loss": 0.0729, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.1974944731024317, | |
| "eval_loss": 0.628384530544281, | |
| "eval_runtime": 75.6565, | |
| "eval_samples_per_second": 15.954, | |
| "eval_steps_per_second": 1.996, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "grad_norm": 4.7338733111246256e-07, | |
| "learning_rate": 2.8506509457135833e-07, | |
| "loss": 0.0476, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.2896094325718497, | |
| "eval_loss": 0.6259528994560242, | |
| "eval_runtime": 75.2674, | |
| "eval_samples_per_second": 16.036, | |
| "eval_steps_per_second": 2.006, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "grad_norm": 28.43704605102539, | |
| "learning_rate": 2.697126013264554e-07, | |
| "loss": 0.0756, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.3817243920412676, | |
| "eval_loss": 0.6102285981178284, | |
| "eval_runtime": 76.1062, | |
| "eval_samples_per_second": 15.859, | |
| "eval_steps_per_second": 1.984, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "grad_norm": 0.0002958567056339234, | |
| "learning_rate": 2.5436010808155247e-07, | |
| "loss": 0.0797, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4738393515106853, | |
| "eval_loss": 0.6023213267326355, | |
| "eval_runtime": 75.1644, | |
| "eval_samples_per_second": 16.058, | |
| "eval_steps_per_second": 2.009, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "grad_norm": 0.0021068258211016655, | |
| "learning_rate": 2.390076148366495e-07, | |
| "loss": 0.0536, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.565954310980103, | |
| "eval_loss": 0.5879408121109009, | |
| "eval_runtime": 76.0196, | |
| "eval_samples_per_second": 15.877, | |
| "eval_steps_per_second": 1.986, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "grad_norm": 4.1152270569000393e-07, | |
| "learning_rate": 2.236551215917465e-07, | |
| "loss": 0.0784, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.658069270449521, | |
| "eval_loss": 0.5879626274108887, | |
| "eval_runtime": 75.6453, | |
| "eval_samples_per_second": 15.956, | |
| "eval_steps_per_second": 1.996, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "grad_norm": 2.4213447913368213e-10, | |
| "learning_rate": 2.083026283468435e-07, | |
| "loss": 0.0703, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.750184229918939, | |
| "eval_loss": 0.5665237307548523, | |
| "eval_runtime": 74.9732, | |
| "eval_samples_per_second": 16.099, | |
| "eval_steps_per_second": 2.014, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "grad_norm": 5.074160799267702e-05, | |
| "learning_rate": 1.9295013510194055e-07, | |
| "loss": 0.0551, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.8422991893883567, | |
| "eval_loss": 0.5670996308326721, | |
| "eval_runtime": 75.902, | |
| "eval_samples_per_second": 15.902, | |
| "eval_steps_per_second": 1.989, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "grad_norm": 6.560769350016926e-08, | |
| "learning_rate": 1.7759764185703757e-07, | |
| "loss": 0.0852, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9344141488577744, | |
| "eval_loss": 0.5694836974143982, | |
| "eval_runtime": 76.048, | |
| "eval_samples_per_second": 15.872, | |
| "eval_steps_per_second": 1.986, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "grad_norm": 4.498223916016286e-06, | |
| "learning_rate": 1.6224514861213458e-07, | |
| "loss": 0.0546, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.026529108327192, | |
| "eval_loss": 0.555758535861969, | |
| "eval_runtime": 74.9986, | |
| "eval_samples_per_second": 16.094, | |
| "eval_steps_per_second": 2.013, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "grad_norm": 273.0765380859375, | |
| "learning_rate": 1.4689265536723165e-07, | |
| "loss": 0.0369, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "eval_loss": 0.5532959699630737, | |
| "eval_runtime": 74.8225, | |
| "eval_samples_per_second": 16.132, | |
| "eval_steps_per_second": 2.018, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "grad_norm": 0.9755889773368835, | |
| "learning_rate": 1.3154016212232866e-07, | |
| "loss": 0.0205, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.210759027266028, | |
| "eval_loss": 0.5498348474502563, | |
| "eval_runtime": 76.4506, | |
| "eval_samples_per_second": 15.788, | |
| "eval_steps_per_second": 1.975, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.3028739867354457, | |
| "grad_norm": 2.7215068340301514, | |
| "learning_rate": 1.1618766887742569e-07, | |
| "loss": 0.0673, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3028739867354457, | |
| "eval_loss": 0.5445749759674072, | |
| "eval_runtime": 74.9944, | |
| "eval_samples_per_second": 16.095, | |
| "eval_steps_per_second": 2.013, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.3949889462048635, | |
| "grad_norm": 0.0005020984681323171, | |
| "learning_rate": 1.0083517563252273e-07, | |
| "loss": 0.0509, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.3949889462048635, | |
| "eval_loss": 0.5434286594390869, | |
| "eval_runtime": 75.0908, | |
| "eval_samples_per_second": 16.074, | |
| "eval_steps_per_second": 2.011, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4871039056742816, | |
| "grad_norm": 0.00025747253675945103, | |
| "learning_rate": 8.548268238761974e-08, | |
| "loss": 0.0447, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.4871039056742816, | |
| "eval_loss": 0.540378749370575, | |
| "eval_runtime": 76.2972, | |
| "eval_samples_per_second": 15.82, | |
| "eval_steps_per_second": 1.979, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.5792188651436994, | |
| "grad_norm": 0.01660293899476528, | |
| "learning_rate": 7.013018914271677e-08, | |
| "loss": 0.0246, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.5792188651436994, | |
| "eval_loss": 0.5360240936279297, | |
| "eval_runtime": 74.959, | |
| "eval_samples_per_second": 16.102, | |
| "eval_steps_per_second": 2.014, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.671333824613117, | |
| "grad_norm": 4.063962055766979e-09, | |
| "learning_rate": 5.47776958978138e-08, | |
| "loss": 0.0395, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.671333824613117, | |
| "eval_loss": 0.533521294593811, | |
| "eval_runtime": 75.1953, | |
| "eval_samples_per_second": 16.052, | |
| "eval_steps_per_second": 2.008, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.7634487840825352, | |
| "grad_norm": 8.483572173645371e-07, | |
| "learning_rate": 3.942520265291083e-08, | |
| "loss": 0.0436, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.7634487840825352, | |
| "eval_loss": 0.5332146286964417, | |
| "eval_runtime": 76.4165, | |
| "eval_samples_per_second": 15.795, | |
| "eval_steps_per_second": 1.976, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.855563743551953, | |
| "grad_norm": 5.3192995147499644e-11, | |
| "learning_rate": 2.407270940800786e-08, | |
| "loss": 0.0398, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.855563743551953, | |
| "eval_loss": 0.5320385098457336, | |
| "eval_runtime": 74.9122, | |
| "eval_samples_per_second": 16.112, | |
| "eval_steps_per_second": 2.016, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.9476787030213707, | |
| "grad_norm": 3.756736993789673, | |
| "learning_rate": 8.720216163104888e-09, | |
| "loss": 0.0427, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.9476787030213707, | |
| "eval_loss": 0.5308398008346558, | |
| "eval_runtime": 75.4523, | |
| "eval_samples_per_second": 15.997, | |
| "eval_steps_per_second": 2.001, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 16284, | |
| "total_flos": 5859024508620360.0, | |
| "train_loss": 0.09254846921924056, | |
| "train_runtime": 12421.3924, | |
| "train_samples_per_second": 2.622, | |
| "train_steps_per_second": 1.311 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16284, | |
| "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": 5859024508620360.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |