| { | |
| "best_metric": 1.9056034088134766, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l55-l/checkpoint-13000", | |
| "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": 0.09209799226376865, | |
| "grad_norm": 0.009597355499863625, | |
| "learning_rate": 4.846503346227052e-05, | |
| "loss": 0.7336, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09209799226376865, | |
| "eval_loss": 2.0780556201934814, | |
| "eval_runtime": 77.1404, | |
| "eval_samples_per_second": 15.634, | |
| "eval_steps_per_second": 1.957, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1841959845275373, | |
| "grad_norm": 0.10049372166395187, | |
| "learning_rate": 4.693006692454105e-05, | |
| "loss": 0.6953, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1841959845275373, | |
| "eval_loss": 2.0782744884490967, | |
| "eval_runtime": 77.8965, | |
| "eval_samples_per_second": 15.482, | |
| "eval_steps_per_second": 1.938, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.27629397679130596, | |
| "grad_norm": 0.0041180383414030075, | |
| "learning_rate": 4.539510038681157e-05, | |
| "loss": 0.6953, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27629397679130596, | |
| "eval_loss": 2.077699899673462, | |
| "eval_runtime": 78.9586, | |
| "eval_samples_per_second": 15.274, | |
| "eval_steps_per_second": 1.912, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3683919690550746, | |
| "grad_norm": 0.29397013783454895, | |
| "learning_rate": 4.386013384908209e-05, | |
| "loss": 0.6932, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3683919690550746, | |
| "eval_loss": 2.0721774101257324, | |
| "eval_runtime": 77.349, | |
| "eval_samples_per_second": 15.592, | |
| "eval_steps_per_second": 1.952, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.46048996131884323, | |
| "grad_norm": 0.9136261343955994, | |
| "learning_rate": 4.232516731135262e-05, | |
| "loss": 0.6907, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46048996131884323, | |
| "eval_loss": 2.079211711883545, | |
| "eval_runtime": 79.3204, | |
| "eval_samples_per_second": 15.204, | |
| "eval_steps_per_second": 1.904, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5525879535826119, | |
| "grad_norm": 4.484213829040527, | |
| "learning_rate": 4.079020077362314e-05, | |
| "loss": 0.7043, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5525879535826119, | |
| "eval_loss": 2.0204577445983887, | |
| "eval_runtime": 78.612, | |
| "eval_samples_per_second": 15.341, | |
| "eval_steps_per_second": 1.921, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6446859458463805, | |
| "grad_norm": 1.1021015644073486, | |
| "learning_rate": 3.9255234235893664e-05, | |
| "loss": 0.7078, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6446859458463805, | |
| "eval_loss": 2.1256344318389893, | |
| "eval_runtime": 78.63, | |
| "eval_samples_per_second": 15.338, | |
| "eval_steps_per_second": 1.92, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7367839381101492, | |
| "grad_norm": 16.864044189453125, | |
| "learning_rate": 3.772026769816418e-05, | |
| "loss": 0.7034, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7367839381101492, | |
| "eval_loss": 2.0000314712524414, | |
| "eval_runtime": 78.6543, | |
| "eval_samples_per_second": 15.333, | |
| "eval_steps_per_second": 1.92, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8288819303739179, | |
| "grad_norm": 20.071943283081055, | |
| "learning_rate": 3.6185301160434705e-05, | |
| "loss": 0.6671, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8288819303739179, | |
| "eval_loss": 2.1607539653778076, | |
| "eval_runtime": 78.6404, | |
| "eval_samples_per_second": 15.336, | |
| "eval_steps_per_second": 1.92, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9209799226376865, | |
| "grad_norm": 17.817607879638672, | |
| "learning_rate": 3.4650334622705225e-05, | |
| "loss": 0.6899, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9209799226376865, | |
| "eval_loss": 2.093959331512451, | |
| "eval_runtime": 78.539, | |
| "eval_samples_per_second": 15.355, | |
| "eval_steps_per_second": 1.923, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.013077914901455, | |
| "grad_norm": 41.0746955871582, | |
| "learning_rate": 3.311536808497575e-05, | |
| "loss": 0.6649, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.013077914901455, | |
| "eval_loss": 1.9776026010513306, | |
| "eval_runtime": 78.5095, | |
| "eval_samples_per_second": 15.361, | |
| "eval_steps_per_second": 1.923, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.1051759071652238, | |
| "grad_norm": 35.426021575927734, | |
| "learning_rate": 3.158040154724627e-05, | |
| "loss": 0.6748, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1051759071652238, | |
| "eval_loss": 1.9851746559143066, | |
| "eval_runtime": 78.5944, | |
| "eval_samples_per_second": 15.345, | |
| "eval_steps_per_second": 1.921, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.1972738994289924, | |
| "grad_norm": 0.8773899078369141, | |
| "learning_rate": 3.0045435009516792e-05, | |
| "loss": 0.6456, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.1972738994289924, | |
| "eval_loss": 1.9981426000595093, | |
| "eval_runtime": 78.4741, | |
| "eval_samples_per_second": 15.368, | |
| "eval_steps_per_second": 1.924, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.289371891692761, | |
| "grad_norm": 19.7215518951416, | |
| "learning_rate": 2.8510468471787316e-05, | |
| "loss": 0.6514, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.289371891692761, | |
| "eval_loss": 1.9636509418487549, | |
| "eval_runtime": 78.4704, | |
| "eval_samples_per_second": 15.369, | |
| "eval_steps_per_second": 1.924, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.3814698839565298, | |
| "grad_norm": 5.96924352645874, | |
| "learning_rate": 2.697550193405784e-05, | |
| "loss": 0.6179, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.3814698839565298, | |
| "eval_loss": 2.0034029483795166, | |
| "eval_runtime": 78.4771, | |
| "eval_samples_per_second": 15.368, | |
| "eval_steps_per_second": 1.924, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.4735678762202984, | |
| "grad_norm": 1.2145048379898071, | |
| "learning_rate": 2.5440535396328363e-05, | |
| "loss": 0.6561, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.4735678762202984, | |
| "eval_loss": 2.2864511013031006, | |
| "eval_runtime": 78.3606, | |
| "eval_samples_per_second": 15.39, | |
| "eval_steps_per_second": 1.927, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.565665868484067, | |
| "grad_norm": 15.816754341125488, | |
| "learning_rate": 2.3905568858598883e-05, | |
| "loss": 0.6328, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.565665868484067, | |
| "eval_loss": 2.4807703495025635, | |
| "eval_runtime": 77.0504, | |
| "eval_samples_per_second": 15.652, | |
| "eval_steps_per_second": 1.96, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.6577638607478358, | |
| "grad_norm": 18.753801345825195, | |
| "learning_rate": 2.2370602320869404e-05, | |
| "loss": 0.7053, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.6577638607478358, | |
| "eval_loss": 2.1609902381896973, | |
| "eval_runtime": 76.5781, | |
| "eval_samples_per_second": 15.749, | |
| "eval_steps_per_second": 1.972, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.7498618530116044, | |
| "grad_norm": 3.3683111667633057, | |
| "learning_rate": 2.0835635783139927e-05, | |
| "loss": 0.6533, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.7498618530116044, | |
| "eval_loss": 1.9845130443572998, | |
| "eval_runtime": 78.8415, | |
| "eval_samples_per_second": 15.297, | |
| "eval_steps_per_second": 1.915, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.841959845275373, | |
| "grad_norm": 5.0674967765808105, | |
| "learning_rate": 1.930066924541045e-05, | |
| "loss": 0.6594, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.841959845275373, | |
| "eval_loss": 1.9894555807113647, | |
| "eval_runtime": 77.1087, | |
| "eval_samples_per_second": 15.64, | |
| "eval_steps_per_second": 1.958, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.9340578375391417, | |
| "grad_norm": 45.641239166259766, | |
| "learning_rate": 1.7765702707680974e-05, | |
| "loss": 0.6597, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9340578375391417, | |
| "eval_loss": 2.0103421211242676, | |
| "eval_runtime": 77.0894, | |
| "eval_samples_per_second": 15.644, | |
| "eval_steps_per_second": 1.959, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.02615582980291, | |
| "grad_norm": 7.026269435882568, | |
| "learning_rate": 1.6230736169951498e-05, | |
| "loss": 0.6435, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.02615582980291, | |
| "eval_loss": 2.058932304382324, | |
| "eval_runtime": 78.5759, | |
| "eval_samples_per_second": 15.348, | |
| "eval_steps_per_second": 1.922, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.118253822066679, | |
| "grad_norm": 11.815095901489258, | |
| "learning_rate": 1.4695769632222017e-05, | |
| "loss": 0.6404, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.118253822066679, | |
| "eval_loss": 2.0517008304595947, | |
| "eval_runtime": 76.9172, | |
| "eval_samples_per_second": 15.679, | |
| "eval_steps_per_second": 1.963, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.2103518143304477, | |
| "grad_norm": 3.2536027431488037, | |
| "learning_rate": 1.316080309449254e-05, | |
| "loss": 0.6348, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.2103518143304477, | |
| "eval_loss": 1.9749338626861572, | |
| "eval_runtime": 78.0449, | |
| "eval_samples_per_second": 15.453, | |
| "eval_steps_per_second": 1.935, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.302449806594216, | |
| "grad_norm": 12.233312606811523, | |
| "learning_rate": 1.1625836556763064e-05, | |
| "loss": 0.6545, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.302449806594216, | |
| "eval_loss": 1.9680836200714111, | |
| "eval_runtime": 76.9869, | |
| "eval_samples_per_second": 15.665, | |
| "eval_steps_per_second": 1.961, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.394547798857985, | |
| "grad_norm": 16.69487953186035, | |
| "learning_rate": 1.0090870019033586e-05, | |
| "loss": 0.6323, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.394547798857985, | |
| "eval_loss": 1.9056034088134766, | |
| "eval_runtime": 78.3412, | |
| "eval_samples_per_second": 15.394, | |
| "eval_steps_per_second": 1.927, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.4866457911217537, | |
| "grad_norm": 11.504305839538574, | |
| "learning_rate": 8.555903481304108e-06, | |
| "loss": 0.6045, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.4866457911217537, | |
| "eval_loss": 1.9470211267471313, | |
| "eval_runtime": 77.1033, | |
| "eval_samples_per_second": 15.641, | |
| "eval_steps_per_second": 1.958, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.578743783385522, | |
| "grad_norm": 188.8672332763672, | |
| "learning_rate": 7.0209369435746295e-06, | |
| "loss": 0.6492, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.578743783385522, | |
| "eval_loss": 2.017714023590088, | |
| "eval_runtime": 77.6236, | |
| "eval_samples_per_second": 15.537, | |
| "eval_steps_per_second": 1.945, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.670841775649291, | |
| "grad_norm": 10.520608901977539, | |
| "learning_rate": 5.485970405845152e-06, | |
| "loss": 0.6512, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.670841775649291, | |
| "eval_loss": 1.948345422744751, | |
| "eval_runtime": 77.8588, | |
| "eval_samples_per_second": 15.49, | |
| "eval_steps_per_second": 1.939, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.7629397679130596, | |
| "grad_norm": 13.767060279846191, | |
| "learning_rate": 3.951003868115676e-06, | |
| "loss": 0.6382, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.7629397679130596, | |
| "eval_loss": 1.9499462842941284, | |
| "eval_runtime": 77.4986, | |
| "eval_samples_per_second": 15.562, | |
| "eval_steps_per_second": 1.948, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.8550377601768284, | |
| "grad_norm": 0.9124027490615845, | |
| "learning_rate": 2.416037330386198e-06, | |
| "loss": 0.6203, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.8550377601768284, | |
| "eval_loss": 1.9528756141662598, | |
| "eval_runtime": 77.9576, | |
| "eval_samples_per_second": 15.47, | |
| "eval_steps_per_second": 1.937, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.947135752440597, | |
| "grad_norm": 8.062443733215332, | |
| "learning_rate": 8.810707926567202e-07, | |
| "loss": 0.6241, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.947135752440597, | |
| "eval_loss": 1.981693148612976, | |
| "eval_runtime": 77.6267, | |
| "eval_samples_per_second": 15.536, | |
| "eval_steps_per_second": 1.945, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 16287, | |
| "total_flos": 5859564146147250.0, | |
| "train_loss": 0.010180538147276201, | |
| "train_runtime": 230.6503, | |
| "train_samples_per_second": 141.214, | |
| "train_steps_per_second": 70.613 | |
| } | |
| ], | |
| "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": 5859564146147250.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |