trainer_state.json

{
  "best_metric": 0.0904286727309227,
  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 260,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.053462374955415726,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.0419,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.2846425771713257,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.0129,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 11.22636604309082,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.0148,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.15062539279460907,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.0796,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.22348184883594513,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1687,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 3.719517946243286,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.1127,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 2.156062126159668,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1401,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.07734036445617676,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0527,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 4.83242654800415,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.1359,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.7723510265350342,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.1215,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.8947368421052632,
      "eval_loss": 0.3576835095882416,
      "eval_runtime": 1.3193,
      "eval_samples_per_second": 100.808,
      "eval_steps_per_second": 12.885,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 4.168729782104492,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0503,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.2171250730752945,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0763,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.038626134395599365,
      "learning_rate": 0.0001,
      "loss": 0.0209,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.0440148189663887,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0375,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.04279002547264099,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0177,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.03838762640953064,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0071,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.5887946486473083,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0074,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.03220641613006592,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0289,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.028111182153224945,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0058,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.027165815234184265,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0054,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.0904286727309227,
      "eval_runtime": 1.2113,
      "eval_samples_per_second": 109.8,
      "eval_steps_per_second": 14.035,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.025491109117865562,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0071,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.024319438263773918,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0051,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 1.6020926237106323,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0076,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.0251007042825222,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0046,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.0962502732872963,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0048,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.02344471961259842,
      "learning_rate": 0.0,
      "loss": 0.0047,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.045078390449858625,
      "train_runtime": 131.1007,
      "train_samples_per_second": 31.548,
      "train_steps_per_second": 1.983
    }
  ],
  "logging_steps": 10,
  "max_steps": 260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.205097416476426e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}