last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004231460911879826,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004231460911879826,
      "grad_norm": 2.2086119651794434,
      "learning_rate": 2e-05,
      "loss": 2.5751,
      "step": 1
    },
    {
      "epoch": 0.0004231460911879826,
      "eval_loss": 2.844357967376709,
      "eval_runtime": 392.816,
      "eval_samples_per_second": 2.533,
      "eval_steps_per_second": 1.268,
      "step": 1
    },
    {
      "epoch": 0.0008462921823759652,
      "grad_norm": 2.279730796813965,
      "learning_rate": 4e-05,
      "loss": 2.8377,
      "step": 2
    },
    {
      "epoch": 0.0012694382735639479,
      "grad_norm": 2.6086931228637695,
      "learning_rate": 6e-05,
      "loss": 2.8085,
      "step": 3
    },
    {
      "epoch": 0.0012694382735639479,
      "eval_loss": 2.829608201980591,
      "eval_runtime": 396.4002,
      "eval_samples_per_second": 2.51,
      "eval_steps_per_second": 1.256,
      "step": 3
    },
    {
      "epoch": 0.0016925843647519305,
      "grad_norm": 2.4624593257904053,
      "learning_rate": 8e-05,
      "loss": 2.6401,
      "step": 4
    },
    {
      "epoch": 0.002115730455939913,
      "grad_norm": 1.834259033203125,
      "learning_rate": 0.0001,
      "loss": 2.195,
      "step": 5
    },
    {
      "epoch": 0.0025388765471278957,
      "grad_norm": 2.621457099914551,
      "learning_rate": 0.00012,
      "loss": 2.7121,
      "step": 6
    },
    {
      "epoch": 0.0025388765471278957,
      "eval_loss": 2.5834829807281494,
      "eval_runtime": 396.4072,
      "eval_samples_per_second": 2.51,
      "eval_steps_per_second": 1.256,
      "step": 6
    },
    {
      "epoch": 0.0029620226383158784,
      "grad_norm": 3.0319173336029053,
      "learning_rate": 0.00014,
      "loss": 2.7119,
      "step": 7
    },
    {
      "epoch": 0.003385168729503861,
      "grad_norm": 2.3886566162109375,
      "learning_rate": 0.00016,
      "loss": 2.4817,
      "step": 8
    },
    {
      "epoch": 0.003808314820691844,
      "grad_norm": 2.3795809745788574,
      "learning_rate": 0.00018,
      "loss": 1.9891,
      "step": 9
    },
    {
      "epoch": 0.003808314820691844,
      "eval_loss": 1.9471328258514404,
      "eval_runtime": 396.3083,
      "eval_samples_per_second": 2.511,
      "eval_steps_per_second": 1.257,
      "step": 9
    },
    {
      "epoch": 0.004231460911879826,
      "grad_norm": 2.5473413467407227,
      "learning_rate": 0.0002,
      "loss": 1.8588,
      "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": 1.318461275897856e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}