trainer_state.json

{
  "best_metric": 0.053506553173065186,
  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-100",
  "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": 1.9774720668792725,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.7886,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.6386513710021973,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3121,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 1.749556541442871,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2196,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.654062032699585,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.3505,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 7.22674036026001,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.2263,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.892793893814087,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2307,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 2.947530508041382,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1071,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.27753958106040955,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0858,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.13430704176425934,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0305,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 4.52810001373291,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0733,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.053506553173065186,
      "eval_runtime": 1.3574,
      "eval_samples_per_second": 97.98,
      "eval_steps_per_second": 12.524,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 2.247769355773926,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.1599,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.35913029313087463,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1081,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5519750118255615,
      "learning_rate": 0.0001,
      "loss": 0.1134,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.08644860982894897,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0423,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.4155566096305847,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0783,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.11766896396875381,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0318,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.20239987969398499,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0284,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.07379453629255295,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0414,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.062465786933898926,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0134,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.06105278804898262,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0254,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.05717790126800537,
      "eval_runtime": 1.3415,
      "eval_samples_per_second": 99.141,
      "eval_steps_per_second": 12.672,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.5096371173858643,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0189,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.05673747509717941,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0114,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.059942204505205154,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0108,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.055583175271749496,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0109,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.060766831040382385,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0106,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.052606601268053055,
      "learning_rate": 0.0,
      "loss": 0.0107,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.12077335517566938,
      "train_runtime": 98.1926,
      "train_samples_per_second": 42.121,
      "train_steps_per_second": 2.648
    }
  ],
  "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
}