checkpoint-3000/trainer_state.json

{
  "best_metric": 0.8782906532287598,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-3000",
  "epoch": 0.33772374197906113,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.056287290329843524,
      "grad_norm": 96.92656707763672,
      "learning_rate": 2.9437127096701565e-07,
      "loss": 0.3748,
      "step": 500
    },
    {
      "epoch": 0.056287290329843524,
      "eval_loss": 1.2226382493972778,
      "eval_runtime": 131.2779,
      "eval_samples_per_second": 15.037,
      "eval_steps_per_second": 1.882,
      "step": 500
    },
    {
      "epoch": 0.11257458065968705,
      "grad_norm": 402.31158447265625,
      "learning_rate": 2.887425419340313e-07,
      "loss": 0.3057,
      "step": 1000
    },
    {
      "epoch": 0.11257458065968705,
      "eval_loss": 1.0702259540557861,
      "eval_runtime": 131.0109,
      "eval_samples_per_second": 15.067,
      "eval_steps_per_second": 1.885,
      "step": 1000
    },
    {
      "epoch": 0.16886187098953057,
      "grad_norm": 515.0911865234375,
      "learning_rate": 2.831138129010469e-07,
      "loss": 0.2239,
      "step": 1500
    },
    {
      "epoch": 0.16886187098953057,
      "eval_loss": 0.9957238435745239,
      "eval_runtime": 134.8377,
      "eval_samples_per_second": 14.64,
      "eval_steps_per_second": 1.832,
      "step": 1500
    },
    {
      "epoch": 0.2251491613193741,
      "grad_norm": 3.1967105865478516,
      "learning_rate": 2.774850838680626e-07,
      "loss": 0.2229,
      "step": 2000
    },
    {
      "epoch": 0.2251491613193741,
      "eval_loss": 0.9504629373550415,
      "eval_runtime": 129.0169,
      "eval_samples_per_second": 15.3,
      "eval_steps_per_second": 1.914,
      "step": 2000
    },
    {
      "epoch": 0.2814364516492176,
      "grad_norm": 276.22998046875,
      "learning_rate": 2.718563548350782e-07,
      "loss": 0.2098,
      "step": 2500
    },
    {
      "epoch": 0.2814364516492176,
      "eval_loss": 0.9006705284118652,
      "eval_runtime": 129.1816,
      "eval_samples_per_second": 15.281,
      "eval_steps_per_second": 1.912,
      "step": 2500
    },
    {
      "epoch": 0.33772374197906113,
      "grad_norm": 1105.317138671875,
      "learning_rate": 2.6622762580209386e-07,
      "loss": 0.1938,
      "step": 3000
    },
    {
      "epoch": 0.33772374197906113,
      "eval_loss": 0.8782906532287598,
      "eval_runtime": 128.6387,
      "eval_samples_per_second": 15.345,
      "eval_steps_per_second": 1.92,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 26649,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1079994570482520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}