trainer_state.json

{
  "best_metric": 0.015739038586616516,
  "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": 2.5359997749328613,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.8503,
      "step": 10
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 1.4746571779251099,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.3911,
      "step": 20
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.8933113813400269,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.186,
      "step": 30
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.3531646728515625,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.2609,
      "step": 40
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.3685688972473145,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.232,
      "step": 50
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.576456308364868,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.2322,
      "step": 60
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.290677011013031,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.1064,
      "step": 70
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.16482844948768616,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1093,
      "step": 80
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.14385782182216644,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.0446,
      "step": 90
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.9377692937850952,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0882,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 1.0,
      "eval_loss": 0.027352238073945045,
      "eval_runtime": 1.2576,
      "eval_samples_per_second": 105.754,
      "eval_steps_per_second": 13.517,
      "step": 100
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 5.423970699310303,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.067,
      "step": 110
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 1.1301406621932983,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0682,
      "step": 120
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.1091245785355568,
      "learning_rate": 0.0001,
      "loss": 0.0753,
      "step": 130
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.11247336864471436,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0614,
      "step": 140
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.10028024762868881,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0745,
      "step": 150
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.11689649522304535,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0373,
      "step": 160
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.11571074277162552,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0173,
      "step": 170
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 9.7699556350708,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0455,
      "step": 180
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.08453625440597534,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0382,
      "step": 190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.06617113202810287,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0571,
      "step": 200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 1.0,
      "eval_loss": 0.015739038586616516,
      "eval_runtime": 1.6025,
      "eval_samples_per_second": 82.995,
      "eval_steps_per_second": 10.608,
      "step": 200
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 0.06209515035152435,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0124,
      "step": 210
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.06629051268100739,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0125,
      "step": 220
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 0.0690915584564209,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0118,
      "step": 230
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.06425461918115616,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0118,
      "step": 240
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.0597422830760479,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0113,
      "step": 250
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0604059100151062,
      "learning_rate": 0.0,
      "loss": 0.0115,
      "step": 260
    },
    {
      "epoch": 4.0,
      "step": 260,
      "total_flos": 3.205097416476426e+17,
      "train_loss": 0.11976414064948375,
      "train_runtime": 91.4122,
      "train_samples_per_second": 45.246,
      "train_steps_per_second": 2.844
    }
  ],
  "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
}