| { | |
| "best_metric": 1.3801825046539307, | |
| "best_model_checkpoint": "./vit-emotion-classification/checkpoint-300", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 2.134533643722534, | |
| "learning_rate": 0.000195, | |
| "loss": 2.0279, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.939251184463501, | |
| "learning_rate": 0.00019, | |
| "loss": 1.7986, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.8353347778320312, | |
| "learning_rate": 0.00018500000000000002, | |
| "loss": 1.6841, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.401373863220215, | |
| "learning_rate": 0.00018, | |
| "loss": 1.5968, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.080655813217163, | |
| "learning_rate": 0.000175, | |
| "loss": 1.3349, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.146280527114868, | |
| "learning_rate": 0.00017, | |
| "loss": 1.2608, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 2.722163438796997, | |
| "learning_rate": 0.000165, | |
| "loss": 1.2728, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.9687062501907349, | |
| "learning_rate": 0.00016, | |
| "loss": 1.1254, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 2.55076265335083, | |
| "learning_rate": 0.000155, | |
| "loss": 0.8877, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.256789445877075, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.8454, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_accuracy": 0.48125, | |
| "eval_loss": 1.4373115301132202, | |
| "eval_runtime": 0.8376, | |
| "eval_samples_per_second": 191.019, | |
| "eval_steps_per_second": 23.877, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 2.4422082901000977, | |
| "learning_rate": 0.000145, | |
| "loss": 0.8337, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.179633140563965, | |
| "learning_rate": 0.00014, | |
| "loss": 0.882, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 1.4560832977294922, | |
| "learning_rate": 0.00013500000000000003, | |
| "loss": 0.5198, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 2.0190017223358154, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.4344, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 3.2591023445129395, | |
| "learning_rate": 0.000125, | |
| "loss": 0.4698, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.652550220489502, | |
| "learning_rate": 0.00012, | |
| "loss": 0.54, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 0.5937463641166687, | |
| "learning_rate": 0.00011499999999999999, | |
| "loss": 0.2737, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 1.0296827554702759, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 0.2402, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 3.007828712463379, | |
| "learning_rate": 0.000105, | |
| "loss": 0.1952, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.936133623123169, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2022, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.55, | |
| "eval_loss": 1.406724214553833, | |
| "eval_runtime": 0.8264, | |
| "eval_samples_per_second": 193.601, | |
| "eval_steps_per_second": 24.2, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "grad_norm": 0.3376343250274658, | |
| "learning_rate": 9.5e-05, | |
| "loss": 0.1086, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 2.637477397918701, | |
| "learning_rate": 9e-05, | |
| "loss": 0.109, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 3.05291485786438, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.0957, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.055624485015869, | |
| "learning_rate": 8e-05, | |
| "loss": 0.136, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.30109065771102905, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0741, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 0.19268257915973663, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0664, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 0.15611745417118073, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.0559, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.21113821864128113, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0599, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "grad_norm": 0.160753071308136, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.049, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.13460688292980194, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0474, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "eval_accuracy": 0.6125, | |
| "eval_loss": 1.3801825046539307, | |
| "eval_runtime": 0.8118, | |
| "eval_samples_per_second": 197.084, | |
| "eval_steps_per_second": 24.635, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 0.13902044296264648, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0452, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.12478330731391907, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0424, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "grad_norm": 0.11532563716173172, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0409, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 0.11974634230136871, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0395, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.1203409880399704, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0393, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.13047201931476593, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0386, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 0.10817253589630127, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0376, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 0.12842506170272827, | |
| "learning_rate": 1e-05, | |
| "loss": 0.037, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "grad_norm": 0.12066592276096344, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0371, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.13028773665428162, | |
| "learning_rate": 0.0, | |
| "loss": 0.0368, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.59375, | |
| "eval_loss": 1.4388375282287598, | |
| "eval_runtime": 0.8121, | |
| "eval_samples_per_second": 197.013, | |
| "eval_steps_per_second": 24.627, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 400, | |
| "total_flos": 4.959754037231616e+17, | |
| "train_loss": 0.4905405020713806, | |
| "train_runtime": 117.1653, | |
| "train_samples_per_second": 54.624, | |
| "train_steps_per_second": 3.414 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 4.959754037231616e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |