trainer_state.json

{
  "best_metric": 0.7983730389308542,
  "best_model_checkpoint": "./sdg-classifier/checkpoint-1076",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1345,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37174721189591076,
      "grad_norm": 3.068047046661377,
      "learning_rate": 1.4370370370370372e-05,
      "loss": 2.5931,
      "step": 100
    },
    {
      "epoch": 0.7434944237918215,
      "grad_norm": 4.889872074127197,
      "learning_rate": 1.897520661157025e-05,
      "loss": 1.2927,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7515398024404416,
      "eval_loss": 0.8947181105613708,
      "eval_runtime": 9.4625,
      "eval_samples_per_second": 909.378,
      "eval_steps_per_second": 7.186,
      "step": 269
    },
    {
      "epoch": 1.1152416356877324,
      "grad_norm": 4.267650604248047,
      "learning_rate": 1.732231404958678e-05,
      "loss": 0.9586,
      "step": 300
    },
    {
      "epoch": 1.486988847583643,
      "grad_norm": 5.622753620147705,
      "learning_rate": 1.566942148760331e-05,
      "loss": 0.8529,
      "step": 400
    },
    {
      "epoch": 1.858736059479554,
      "grad_norm": 4.57409143447876,
      "learning_rate": 1.4016528925619836e-05,
      "loss": 0.7953,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7795467751307379,
      "eval_loss": 0.7700016498565674,
      "eval_runtime": 9.1997,
      "eval_samples_per_second": 935.353,
      "eval_steps_per_second": 7.392,
      "step": 538
    },
    {
      "epoch": 2.2304832713754648,
      "grad_norm": 5.621212005615234,
      "learning_rate": 1.2363636363636364e-05,
      "loss": 0.7169,
      "step": 600
    },
    {
      "epoch": 2.6022304832713754,
      "grad_norm": 7.005183696746826,
      "learning_rate": 1.0710743801652894e-05,
      "loss": 0.666,
      "step": 700
    },
    {
      "epoch": 2.973977695167286,
      "grad_norm": 5.196578025817871,
      "learning_rate": 9.057851239669422e-06,
      "loss": 0.6549,
      "step": 800
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7937245787332946,
      "eval_loss": 0.7241168022155762,
      "eval_runtime": 9.2798,
      "eval_samples_per_second": 927.284,
      "eval_steps_per_second": 7.328,
      "step": 807
    },
    {
      "epoch": 3.345724907063197,
      "grad_norm": 4.715315341949463,
      "learning_rate": 7.40495867768595e-06,
      "loss": 0.5591,
      "step": 900
    },
    {
      "epoch": 3.717472118959108,
      "grad_norm": 7.226228713989258,
      "learning_rate": 5.7520661157024795e-06,
      "loss": 0.5658,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7983730389308542,
      "eval_loss": 0.7134895920753479,
      "eval_runtime": 9.1183,
      "eval_samples_per_second": 943.703,
      "eval_steps_per_second": 7.458,
      "step": 1076
    },
    {
      "epoch": 4.089219330855019,
      "grad_norm": 7.13743782043457,
      "learning_rate": 4.099173553719009e-06,
      "loss": 0.542,
      "step": 1100
    },
    {
      "epoch": 4.4609665427509295,
      "grad_norm": 6.263808727264404,
      "learning_rate": 2.4462809917355375e-06,
      "loss": 0.4922,
      "step": 1200
    },
    {
      "epoch": 4.83271375464684,
      "grad_norm": 7.6533989906311035,
      "learning_rate": 7.933884297520662e-07,
      "loss": 0.4799,
      "step": 1300
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7940732132481115,
      "eval_loss": 0.7142194509506226,
      "eval_runtime": 9.2321,
      "eval_samples_per_second": 932.07,
      "eval_steps_per_second": 7.366,
      "step": 1345
    },
    {
      "epoch": 5.0,
      "step": 1345,
      "total_flos": 4.52871045169152e+16,
      "train_loss": 0.846687467213457,
      "train_runtime": 657.9188,
      "train_samples_per_second": 261.582,
      "train_steps_per_second": 2.044
    }
  ],
  "logging_steps": 100,
  "max_steps": 1345,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 5000.0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.52871045169152e+16,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}