trainer_state.json

{
  "best_metric": 0.022380338981747627,
  "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": 1.6479884386062622,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8396,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 1.509365200996399,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3393,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.691885769367218,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.1684,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 6.484596252441406,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2333,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.8088409900665283,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.2255,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 6.121501922607422,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2669,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.24062782526016235,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.099,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.21734175086021423,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1085,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.19399584829807281,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0505,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.30884402990341187,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0905,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.9699248120300752,
      "eval_loss": 0.0948573499917984,
      "eval_runtime": 1.2828,
      "eval_samples_per_second": 103.681,
      "eval_steps_per_second": 13.252,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 3.9459798336029053,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0638,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.5659621357917786,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.1135,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.21752521395683289,
      "learning_rate": 0.0001,
      "loss": 0.0562,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.09279836714267731,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0614,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 3.6815249919891357,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0821,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.07715422660112381,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0392,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 1.1431068181991577,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0216,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 3.275928258895874,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0453,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.09814456850290298,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0278,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.060878027230501175,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0149,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.9924812030075187,
      "eval_loss": 0.022380338981747627,
      "eval_runtime": 1.2374,
      "eval_samples_per_second": 107.485,
      "eval_steps_per_second": 13.739,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.06566526740789413,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0114,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.18202857673168182,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0111,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.05648389086127281,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0105,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.06603652983903885,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0102,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.054385650902986526,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0104,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05062974616885185,
      "learning_rate": 0.0,
      "loss": 0.0101,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11581076618570547,
      "train_runtime": 97.3616,
      "train_samples_per_second": 42.481,
      "train_steps_per_second": 2.67
    }
  ],
  "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
}