| { | |
| "best_metric": 0.42857142857142855, | |
| "best_model_checkpoint": "Tianjiao-Yu/videomae-huge/checkpoint-70", | |
| "epoch": 5.102564102564102, | |
| "eval_steps": 500, | |
| "global_step": 78, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.125e-05, | |
| "loss": 2.66, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.8571428571428576e-05, | |
| "loss": 2.6392, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.25, | |
| "eval_loss": 2.5779731273651123, | |
| "eval_runtime": 7.9027, | |
| "eval_samples_per_second": 3.543, | |
| "eval_steps_per_second": 0.506, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.5921, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 2.2172, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.785714285714286e-05, | |
| "loss": 1.994, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_accuracy": 0.39285714285714285, | |
| "eval_loss": 2.399493932723999, | |
| "eval_runtime": 7.938, | |
| "eval_samples_per_second": 3.527, | |
| "eval_steps_per_second": 0.504, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 3.428571428571429e-05, | |
| "loss": 1.947, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.071428571428572e-05, | |
| "loss": 1.4849, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 1.6374, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_accuracy": 0.39285714285714285, | |
| "eval_loss": 2.3010246753692627, | |
| "eval_runtime": 7.9782, | |
| "eval_samples_per_second": 3.51, | |
| "eval_steps_per_second": 0.501, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 2.357142857142857e-05, | |
| "loss": 1.3647, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 2e-05, | |
| "loss": 1.216, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 1.642857142857143e-05, | |
| "loss": 1.124, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "eval_accuracy": 0.39285714285714285, | |
| "eval_loss": 2.2242484092712402, | |
| "eval_runtime": 7.7889, | |
| "eval_samples_per_second": 3.595, | |
| "eval_steps_per_second": 0.514, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 1.0848, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 9.285714285714286e-06, | |
| "loss": 0.9794, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.9569, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 2.182518482208252, | |
| "eval_runtime": 7.9247, | |
| "eval_samples_per_second": 3.533, | |
| "eval_steps_per_second": 0.505, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 2.142857142857143e-06, | |
| "loss": 0.8862, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 2.1734211444854736, | |
| "eval_runtime": 7.8943, | |
| "eval_samples_per_second": 3.547, | |
| "eval_steps_per_second": 0.507, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "step": 78, | |
| "total_flos": 7.713965251203564e+17, | |
| "train_loss": 1.6192821661631267, | |
| "train_runtime": 267.6762, | |
| "train_samples_per_second": 2.331, | |
| "train_steps_per_second": 0.291 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 2.0714259147644043, | |
| "eval_runtime": 4.1878, | |
| "eval_samples_per_second": 3.343, | |
| "eval_steps_per_second": 0.478, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 2.0714259147644043, | |
| "eval_runtime": 3.8707, | |
| "eval_samples_per_second": 3.617, | |
| "eval_steps_per_second": 0.517, | |
| "step": 78 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 78, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 7.713965251203564e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |