trainer_state.json

{
  "best_metric": 0.05081319808959961,
  "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": 1.4000637531280518,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8277,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 1.564321756362915,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3406,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.7980456352233887,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.2452,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 1.2301970720291138,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.222,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.8124799728393555,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.1457,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.9397706389427185,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.1581,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 4.101640224456787,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.2372,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.16402725875377655,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.0753,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.1545068770647049,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0828,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 2.834233522415161,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0655,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9774436090225563,
      "eval_loss": 0.09004394710063934,
      "eval_runtime": 1.7108,
      "eval_samples_per_second": 77.741,
      "eval_steps_per_second": 9.937,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.39476487040519714,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0968,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 2.3521738052368164,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1008,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.230108618736267,
      "learning_rate": 0.0001,
      "loss": 0.0803,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 1.884257197380066,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0488,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.30717819929122925,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0202,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.07915246486663818,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0233,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.07778707146644592,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0353,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.09130260348320007,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0222,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.0661533921957016,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0141,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.09436892718076706,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0184,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9849624060150376,
      "eval_loss": 0.05081319808959961,
      "eval_runtime": 1.2254,
      "eval_samples_per_second": 108.54,
      "eval_steps_per_second": 13.874,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.06286436319351196,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0295,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.058415208011865616,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0122,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.06178402155637741,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0109,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.062430400401353836,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.011,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.05799039453268051,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0111,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.347285509109497,
      "learning_rate": 0.0,
      "loss": 0.0123,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11335522171396475,
      "train_runtime": 88.885,
      "train_samples_per_second": 46.532,
      "train_steps_per_second": 2.925
    }
  ],
  "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
}