trainer_state.json

{
  "best_metric": 0.01560910977423191,
  "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": 1.833648920059204,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8819,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 1.644918441772461,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3892,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 4.5665059089660645,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2533,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.2684237957000732,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2013,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.3962904214859009,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1772,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 5.1289777755737305,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2362,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.1202151775360107,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.0762,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.15739357471466064,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0819,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.14268840849399567,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0581,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 1.1018327474594116,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0621,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.05534913018345833,
      "eval_runtime": 1.2574,
      "eval_samples_per_second": 105.77,
      "eval_steps_per_second": 13.519,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.9867056608200073,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.1224,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.5382664799690247,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0767,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.747385025024414,
      "learning_rate": 0.0001,
      "loss": 0.0618,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 3.467733383178711,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0932,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.1587020605802536,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0236,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.10661860555410385,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0453,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.1051967591047287,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0416,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.0817098394036293,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0156,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.07392418384552002,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0145,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.07026856392621994,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.015,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01560910977423191,
      "eval_runtime": 1.7933,
      "eval_samples_per_second": 74.166,
      "eval_steps_per_second": 9.48,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.06522107124328613,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0131,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.06842605024576187,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0128,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.0690511018037796,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0123,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.06551249325275421,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0121,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.0707441195845604,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0118,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0675133541226387,
      "learning_rate": 0.0,
      "loss": 0.0151,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11554959192872047,
      "train_runtime": 82.7629,
      "train_samples_per_second": 49.974,
      "train_steps_per_second": 3.142
    }
  ],
  "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
}