last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00023621256768966393,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3621256768966393e-05,
      "grad_norm": 0.8800978660583496,
      "learning_rate": 2e-05,
      "loss": 7.4129,
      "step": 1
    },
    {
      "epoch": 2.3621256768966393e-05,
      "eval_loss": 1.6822291612625122,
      "eval_runtime": 1333.2677,
      "eval_samples_per_second": 13.37,
      "eval_steps_per_second": 6.685,
      "step": 1
    },
    {
      "epoch": 4.7242513537932786e-05,
      "grad_norm": 0.8495243191719055,
      "learning_rate": 4e-05,
      "loss": 5.8307,
      "step": 2
    },
    {
      "epoch": 7.086377030689917e-05,
      "grad_norm": 0.8420003652572632,
      "learning_rate": 6e-05,
      "loss": 6.893,
      "step": 3
    },
    {
      "epoch": 7.086377030689917e-05,
      "eval_loss": 1.6812106370925903,
      "eval_runtime": 1336.9399,
      "eval_samples_per_second": 13.333,
      "eval_steps_per_second": 6.667,
      "step": 3
    },
    {
      "epoch": 9.448502707586557e-05,
      "grad_norm": 0.9201523065567017,
      "learning_rate": 8e-05,
      "loss": 6.5537,
      "step": 4
    },
    {
      "epoch": 0.00011810628384483197,
      "grad_norm": 0.8098495006561279,
      "learning_rate": 0.0001,
      "loss": 7.3419,
      "step": 5
    },
    {
      "epoch": 0.00014172754061379835,
      "grad_norm": 1.112677812576294,
      "learning_rate": 0.00012,
      "loss": 7.3867,
      "step": 6
    },
    {
      "epoch": 0.00014172754061379835,
      "eval_loss": 1.644856333732605,
      "eval_runtime": 1337.5689,
      "eval_samples_per_second": 13.327,
      "eval_steps_per_second": 6.664,
      "step": 6
    },
    {
      "epoch": 0.00016534879738276475,
      "grad_norm": 1.1206536293029785,
      "learning_rate": 0.00014,
      "loss": 6.9054,
      "step": 7
    },
    {
      "epoch": 0.00018897005415173114,
      "grad_norm": 0.9806450605392456,
      "learning_rate": 0.00016,
      "loss": 6.1312,
      "step": 8
    },
    {
      "epoch": 0.00021259131092069754,
      "grad_norm": 1.225191593170166,
      "learning_rate": 0.00018,
      "loss": 7.6522,
      "step": 9
    },
    {
      "epoch": 0.00021259131092069754,
      "eval_loss": 1.5392491817474365,
      "eval_runtime": 1337.6222,
      "eval_samples_per_second": 13.327,
      "eval_steps_per_second": 6.663,
      "step": 9
    },
    {
      "epoch": 0.00023621256768966393,
      "grad_norm": 1.4865165948867798,
      "learning_rate": 0.0002,
      "loss": 6.1353,
      "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": 6196217134448640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}