{
  "best_metric": 0.926975429058075,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-11000",
  "epoch": 2.02615582980291,
  "eval_steps": 500,
  "global_step": 11000,
  "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
    },
    {
      "epoch": 0.8288819303739179,
      "grad_norm": 1.6036031246185303,
      "learning_rate": 3.9638975870326026e-08,
      "loss": 0.307,
      "step": 4500
    },
    {
      "epoch": 0.8288819303739179,
      "eval_loss": 1.0728861093521118,
      "eval_runtime": 78.4032,
      "eval_samples_per_second": 15.382,
      "eval_steps_per_second": 1.926,
      "step": 4500
    },
    {
      "epoch": 0.9209799226376865,
      "grad_norm": 11.990758895874023,
      "learning_rate": 3.8487750967028915e-08,
      "loss": 0.2764,
      "step": 5000
    },
    {
      "epoch": 0.9209799226376865,
      "eval_loss": 1.0523593425750732,
      "eval_runtime": 78.3726,
      "eval_samples_per_second": 15.388,
      "eval_steps_per_second": 1.927,
      "step": 5000
    },
    {
      "epoch": 1.013077914901455,
      "grad_norm": 43.865562438964844,
      "learning_rate": 3.733652606373181e-08,
      "loss": 0.2456,
      "step": 5500
    },
    {
      "epoch": 1.013077914901455,
      "eval_loss": 1.0375442504882812,
      "eval_runtime": 78.2473,
      "eval_samples_per_second": 15.413,
      "eval_steps_per_second": 1.93,
      "step": 5500
    },
    {
      "epoch": 1.1051759071652238,
      "grad_norm": 105.91036987304688,
      "learning_rate": 3.61853011604347e-08,
      "loss": 0.2642,
      "step": 6000
    },
    {
      "epoch": 1.1051759071652238,
      "eval_loss": 1.0233356952667236,
      "eval_runtime": 78.3135,
      "eval_samples_per_second": 15.4,
      "eval_steps_per_second": 1.928,
      "step": 6000
    },
    {
      "epoch": 1.1972738994289924,
      "grad_norm": 221.4896240234375,
      "learning_rate": 3.503407625713759e-08,
      "loss": 0.2066,
      "step": 6500
    },
    {
      "epoch": 1.1972738994289924,
      "eval_loss": 1.0104238986968994,
      "eval_runtime": 78.6485,
      "eval_samples_per_second": 15.334,
      "eval_steps_per_second": 1.92,
      "step": 6500
    },
    {
      "epoch": 1.289371891692761,
      "grad_norm": 270.53253173828125,
      "learning_rate": 3.3882851353840485e-08,
      "loss": 0.2376,
      "step": 7000
    },
    {
      "epoch": 1.289371891692761,
      "eval_loss": 0.998353898525238,
      "eval_runtime": 78.5502,
      "eval_samples_per_second": 15.353,
      "eval_steps_per_second": 1.922,
      "step": 7000
    },
    {
      "epoch": 1.3814698839565298,
      "grad_norm": 402.9339904785156,
      "learning_rate": 3.2731626450543374e-08,
      "loss": 0.1931,
      "step": 7500
    },
    {
      "epoch": 1.3814698839565298,
      "eval_loss": 0.9887453317642212,
      "eval_runtime": 78.4833,
      "eval_samples_per_second": 15.366,
      "eval_steps_per_second": 1.924,
      "step": 7500
    },
    {
      "epoch": 1.4735678762202984,
      "grad_norm": 0.006355441175401211,
      "learning_rate": 3.158040154724626e-08,
      "loss": 0.2163,
      "step": 8000
    },
    {
      "epoch": 1.4735678762202984,
      "eval_loss": 0.9767189621925354,
      "eval_runtime": 78.5499,
      "eval_samples_per_second": 15.353,
      "eval_steps_per_second": 1.922,
      "step": 8000
    },
    {
      "epoch": 1.565665868484067,
      "grad_norm": 0.12982100248336792,
      "learning_rate": 3.042917664394916e-08,
      "loss": 0.1903,
      "step": 8500
    },
    {
      "epoch": 1.565665868484067,
      "eval_loss": 0.9664921760559082,
      "eval_runtime": 78.6832,
      "eval_samples_per_second": 15.327,
      "eval_steps_per_second": 1.919,
      "step": 8500
    },
    {
      "epoch": 1.6577638607478358,
      "grad_norm": 173.31788635253906,
      "learning_rate": 2.927795174065205e-08,
      "loss": 0.2069,
      "step": 9000
    },
    {
      "epoch": 1.6577638607478358,
      "eval_loss": 0.9571623206138611,
      "eval_runtime": 78.544,
      "eval_samples_per_second": 15.354,
      "eval_steps_per_second": 1.922,
      "step": 9000
    },
    {
      "epoch": 1.7498618530116044,
      "grad_norm": 358.7403259277344,
      "learning_rate": 2.8126726837354947e-08,
      "loss": 0.2093,
      "step": 9500
    },
    {
      "epoch": 1.7498618530116044,
      "eval_loss": 0.9496769905090332,
      "eval_runtime": 78.3124,
      "eval_samples_per_second": 15.4,
      "eval_steps_per_second": 1.928,
      "step": 9500
    },
    {
      "epoch": 1.841959845275373,
      "grad_norm": 435.9449768066406,
      "learning_rate": 2.6975501934057836e-08,
      "loss": 0.2523,
      "step": 10000
    },
    {
      "epoch": 1.841959845275373,
      "eval_loss": 0.9420022368431091,
      "eval_runtime": 78.3702,
      "eval_samples_per_second": 15.388,
      "eval_steps_per_second": 1.927,
      "step": 10000
    },
    {
      "epoch": 1.9340578375391417,
      "grad_norm": 23.49118423461914,
      "learning_rate": 2.582427703076073e-08,
      "loss": 0.2127,
      "step": 10500
    },
    {
      "epoch": 1.9340578375391417,
      "eval_loss": 0.9328628182411194,
      "eval_runtime": 76.6904,
      "eval_samples_per_second": 15.726,
      "eval_steps_per_second": 1.969,
      "step": 10500
    },
    {
      "epoch": 2.02615582980291,
      "grad_norm": 0.4029064476490021,
      "learning_rate": 2.467305212746362e-08,
      "loss": 0.1968,
      "step": 11000
    },
    {
      "epoch": 2.02615582980291,
      "eval_loss": 0.926975429058075,
      "eval_runtime": 76.7161,
      "eval_samples_per_second": 15.72,
      "eval_steps_per_second": 1.968,
      "step": 11000
    }
  ],
  "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": 3957701622211260.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}