trainer_state.json

{
  "best_metric": 0.013332394883036613,
  "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": 1.4326872825622559,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.7835,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 1.9304112195968628,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.356,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 3.0517399311065674,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2107,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 3.5913784503936768,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2222,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 4.897500038146973,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.2676,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 3.6991374492645264,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.3677,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.5888941287994385,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.0946,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 5.136747360229492,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1049,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 4.133493900299072,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0366,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 5.138324737548828,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0677,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.05553809180855751,
      "eval_runtime": 1.2866,
      "eval_samples_per_second": 103.375,
      "eval_steps_per_second": 13.213,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.1761542111635208,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.1063,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 3.5044777393341064,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1599,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.8694207668304443,
      "learning_rate": 0.0001,
      "loss": 0.1102,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.2651481032371521,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0379,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 6.802685260772705,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0465,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.1000690683722496,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0511,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.7406809329986572,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0206,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.2656266689300537,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0656,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.1580563485622406,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0238,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.06784767657518387,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0148,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 1.0,
      "eval_loss": 0.013332394883036613,
      "eval_runtime": 1.2533,
      "eval_samples_per_second": 106.118,
      "eval_steps_per_second": 13.564,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.059637218713760376,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.039,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.0609792061150074,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.012,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.06473277509212494,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0117,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.06173160299658775,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0112,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.05901215225458145,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0109,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05926782637834549,
      "learning_rate": 0.0,
      "loss": 0.0163,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.12496758367006595,
      "train_runtime": 90.9233,
      "train_samples_per_second": 45.489,
      "train_steps_per_second": 2.86
    }
  ],
  "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
}