last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008618275052248292,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.618275052248293e-05,
      "grad_norm": 1.5035656690597534,
      "learning_rate": 2e-05,
      "loss": 1.7574,
      "step": 1
    },
    {
      "epoch": 8.618275052248293e-05,
      "eval_loss": NaN,
      "eval_runtime": 196.2611,
      "eval_samples_per_second": 24.895,
      "eval_steps_per_second": 12.448,
      "step": 1
    },
    {
      "epoch": 0.00017236550104496586,
      "grad_norm": 1.076463222503662,
      "learning_rate": 4e-05,
      "loss": 1.6179,
      "step": 2
    },
    {
      "epoch": 0.0002585482515674488,
      "grad_norm": 1.1540504693984985,
      "learning_rate": 6e-05,
      "loss": 1.3239,
      "step": 3
    },
    {
      "epoch": 0.0002585482515674488,
      "eval_loss": NaN,
      "eval_runtime": 196.1042,
      "eval_samples_per_second": 24.915,
      "eval_steps_per_second": 12.458,
      "step": 3
    },
    {
      "epoch": 0.0003447310020899317,
      "grad_norm": 1.6838661432266235,
      "learning_rate": 8e-05,
      "loss": 1.7573,
      "step": 4
    },
    {
      "epoch": 0.0004309137526124146,
      "grad_norm": 1.3160624504089355,
      "learning_rate": 0.0001,
      "loss": 1.49,
      "step": 5
    },
    {
      "epoch": 0.0005170965031348976,
      "grad_norm": 1.4394317865371704,
      "learning_rate": 0.00012,
      "loss": 1.7809,
      "step": 6
    },
    {
      "epoch": 0.0005170965031348976,
      "eval_loss": NaN,
      "eval_runtime": 196.1687,
      "eval_samples_per_second": 24.907,
      "eval_steps_per_second": 12.454,
      "step": 6
    },
    {
      "epoch": 0.0006032792536573804,
      "grad_norm": 1.0919623374938965,
      "learning_rate": 0.00014,
      "loss": 1.0546,
      "step": 7
    },
    {
      "epoch": 0.0006894620041798634,
      "grad_norm": 1.4632925987243652,
      "learning_rate": 0.00016,
      "loss": 1.6939,
      "step": 8
    },
    {
      "epoch": 0.0007756447547023463,
      "grad_norm": 2.3321034908294678,
      "learning_rate": 0.00018,
      "loss": 2.6489,
      "step": 9
    },
    {
      "epoch": 0.0007756447547023463,
      "eval_loss": NaN,
      "eval_runtime": 196.1766,
      "eval_samples_per_second": 24.906,
      "eval_steps_per_second": 12.453,
      "step": 9
    },
    {
      "epoch": 0.0008618275052248292,
      "grad_norm": 0.6794093251228333,
      "learning_rate": 0.0002,
      "loss": 0.6383,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1189070564229120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}