checkpoint-4000/trainer_state.json

{
  "best_metric": 1.0954023599624634,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-4000",
  "epoch": 0.7367839381101492,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09209799226376865,
      "grad_norm": 0.0014697719598188996,
      "learning_rate": 4.884877509670289e-08,
      "loss": 0.4588,
      "step": 500
    },
    {
      "epoch": 0.09209799226376865,
      "eval_loss": 1.4194883108139038,
      "eval_runtime": 78.7703,
      "eval_samples_per_second": 15.31,
      "eval_steps_per_second": 1.917,
      "step": 500
    },
    {
      "epoch": 0.1841959845275373,
      "grad_norm": 301.76043701171875,
      "learning_rate": 4.769755019340578e-08,
      "loss": 0.4255,
      "step": 1000
    },
    {
      "epoch": 0.1841959845275373,
      "eval_loss": 1.3416553735733032,
      "eval_runtime": 77.0051,
      "eval_samples_per_second": 15.661,
      "eval_steps_per_second": 1.961,
      "step": 1000
    },
    {
      "epoch": 0.27629397679130596,
      "grad_norm": 35.28853225708008,
      "learning_rate": 4.654632529010867e-08,
      "loss": 0.3724,
      "step": 1500
    },
    {
      "epoch": 0.27629397679130596,
      "eval_loss": 1.2872973680496216,
      "eval_runtime": 77.5439,
      "eval_samples_per_second": 15.552,
      "eval_steps_per_second": 1.947,
      "step": 1500
    },
    {
      "epoch": 0.3683919690550746,
      "grad_norm": 338.9349060058594,
      "learning_rate": 4.539510038681156e-08,
      "loss": 0.3251,
      "step": 2000
    },
    {
      "epoch": 0.3683919690550746,
      "eval_loss": 1.234910249710083,
      "eval_runtime": 78.7715,
      "eval_samples_per_second": 15.31,
      "eval_steps_per_second": 1.917,
      "step": 2000
    },
    {
      "epoch": 0.46048996131884323,
      "grad_norm": 325.85101318359375,
      "learning_rate": 4.4243875483514457e-08,
      "loss": 0.3308,
      "step": 2500
    },
    {
      "epoch": 0.46048996131884323,
      "eval_loss": 1.1944907903671265,
      "eval_runtime": 79.0397,
      "eval_samples_per_second": 15.258,
      "eval_steps_per_second": 1.91,
      "step": 2500
    },
    {
      "epoch": 0.5525879535826119,
      "grad_norm": 100.85855102539062,
      "learning_rate": 4.309265058021735e-08,
      "loss": 0.3017,
      "step": 3000
    },
    {
      "epoch": 0.5525879535826119,
      "eval_loss": 1.1593009233474731,
      "eval_runtime": 77.6354,
      "eval_samples_per_second": 15.534,
      "eval_steps_per_second": 1.945,
      "step": 3000
    },
    {
      "epoch": 0.6446859458463805,
      "grad_norm": 8.865598678588867,
      "learning_rate": 4.194142567692024e-08,
      "loss": 0.2962,
      "step": 3500
    },
    {
      "epoch": 0.6446859458463805,
      "eval_loss": 1.1259427070617676,
      "eval_runtime": 78.2715,
      "eval_samples_per_second": 15.408,
      "eval_steps_per_second": 1.929,
      "step": 3500
    },
    {
      "epoch": 0.7367839381101492,
      "grad_norm": 4.393447399139404,
      "learning_rate": 4.079020077362314e-08,
      "loss": 0.2919,
      "step": 4000
    },
    {
      "epoch": 0.7367839381101492,
      "eval_loss": 1.0954023599624634,
      "eval_runtime": 78.5366,
      "eval_samples_per_second": 15.356,
      "eval_steps_per_second": 1.923,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 21716,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 1439752921742520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}