trainer_state.json

{
  "best_metric": 1.0,
  "best_model_checkpoint": "videomae-base-finetuned-Risky-situations/checkpoint-25",
  "epoch": 4.2,
  "eval_steps": 500,
  "global_step": 125,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 14.0280122756958,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.7118,
      "step": 10
    },
    {
      "epoch": 0.16,
      "grad_norm": 22.5030574798584,
      "learning_rate": 4.6875e-05,
      "loss": 0.7248,
      "step": 20
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 1.0,
      "eval_loss": 0.3613555431365967,
      "eval_runtime": 13.5011,
      "eval_samples_per_second": 0.889,
      "eval_steps_per_second": 0.444,
      "step": 25
    },
    {
      "epoch": 1.04,
      "grad_norm": 7.6639909744262695,
      "learning_rate": 4.2410714285714285e-05,
      "loss": 0.4691,
      "step": 30
    },
    {
      "epoch": 1.12,
      "grad_norm": 1.105189323425293,
      "learning_rate": 3.794642857142857e-05,
      "loss": 0.148,
      "step": 40
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.10660043358802795,
      "learning_rate": 3.348214285714286e-05,
      "loss": 0.829,
      "step": 50
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.9166666666666666,
      "eval_loss": 0.22277389466762543,
      "eval_runtime": 13.4172,
      "eval_samples_per_second": 0.894,
      "eval_steps_per_second": 0.447,
      "step": 50
    },
    {
      "epoch": 2.08,
      "grad_norm": 144.1703338623047,
      "learning_rate": 2.9017857142857146e-05,
      "loss": 0.3704,
      "step": 60
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.0353066623210907,
      "learning_rate": 2.455357142857143e-05,
      "loss": 0.3176,
      "step": 70
    },
    {
      "epoch": 2.2,
      "eval_accuracy": 1.0,
      "eval_loss": 0.006112185772508383,
      "eval_runtime": 14.0771,
      "eval_samples_per_second": 0.852,
      "eval_steps_per_second": 0.426,
      "step": 75
    },
    {
      "epoch": 3.04,
      "grad_norm": 0.020980490371584892,
      "learning_rate": 2.0089285714285717e-05,
      "loss": 0.3022,
      "step": 80
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.20112857222557068,
      "learning_rate": 1.5625e-05,
      "loss": 0.252,
      "step": 90
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.01630263216793537,
      "learning_rate": 1.1160714285714287e-05,
      "loss": 0.0027,
      "step": 100
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0014307590899989009,
      "eval_runtime": 13.5819,
      "eval_samples_per_second": 0.884,
      "eval_steps_per_second": 0.442,
      "step": 100
    },
    {
      "epoch": 4.08,
      "grad_norm": 0.049904562532901764,
      "learning_rate": 6.696428571428572e-06,
      "loss": 0.0018,
      "step": 110
    },
    {
      "epoch": 4.16,
      "grad_norm": 0.013597003184258938,
      "learning_rate": 2.2321428571428573e-06,
      "loss": 0.0011,
      "step": 120
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0012231767177581787,
      "eval_runtime": 15.5498,
      "eval_samples_per_second": 0.772,
      "eval_steps_per_second": 0.386,
      "step": 125
    },
    {
      "epoch": 4.2,
      "step": 125,
      "total_flos": 3.11516147367936e+17,
      "train_loss": 0.3304869447611272,
      "train_runtime": 473.2289,
      "train_samples_per_second": 0.528,
      "train_steps_per_second": 0.264
    },
    {
      "epoch": 4.2,
      "eval_accuracy": 0.8333333333333334,
      "eval_loss": 0.565607488155365,
      "eval_runtime": 20.5092,
      "eval_samples_per_second": 0.585,
      "eval_steps_per_second": 0.293,
      "step": 125
    }
  ],
  "logging_steps": 10,
  "max_steps": 125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "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": 3.11516147367936e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}