last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5333333333333333,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 15.137728691101074,
      "learning_rate": 2e-05,
      "loss": 6.5027,
      "step": 1
    },
    {
      "epoch": 0.05333333333333334,
      "eval_loss": 7.445281505584717,
      "eval_runtime": 0.9089,
      "eval_samples_per_second": 8.802,
      "eval_steps_per_second": 4.401,
      "step": 1
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 13.175712585449219,
      "learning_rate": 4e-05,
      "loss": 7.4042,
      "step": 2
    },
    {
      "epoch": 0.16,
      "grad_norm": 16.013059616088867,
      "learning_rate": 6e-05,
      "loss": 7.8371,
      "step": 3
    },
    {
      "epoch": 0.16,
      "eval_loss": 7.387241840362549,
      "eval_runtime": 0.2132,
      "eval_samples_per_second": 37.524,
      "eval_steps_per_second": 18.762,
      "step": 3
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 24.238176345825195,
      "learning_rate": 8e-05,
      "loss": 8.4837,
      "step": 4
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 20.77908706665039,
      "learning_rate": 0.0001,
      "loss": 7.1379,
      "step": 5
    },
    {
      "epoch": 0.32,
      "grad_norm": 15.75521469116211,
      "learning_rate": 0.00012,
      "loss": 5.6818,
      "step": 6
    },
    {
      "epoch": 0.32,
      "eval_loss": 5.2360687255859375,
      "eval_runtime": 0.2203,
      "eval_samples_per_second": 36.318,
      "eval_steps_per_second": 18.159,
      "step": 6
    },
    {
      "epoch": 0.37333333333333335,
      "grad_norm": 28.851076126098633,
      "learning_rate": 0.00014,
      "loss": 5.5918,
      "step": 7
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 17.395139694213867,
      "learning_rate": 0.00016,
      "loss": 4.411,
      "step": 8
    },
    {
      "epoch": 0.48,
      "grad_norm": 13.205751419067383,
      "learning_rate": 0.00018,
      "loss": 2.9019,
      "step": 9
    },
    {
      "epoch": 0.48,
      "eval_loss": 1.100304126739502,
      "eval_runtime": 0.2364,
      "eval_samples_per_second": 33.834,
      "eval_steps_per_second": 16.917,
      "step": 9
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 6.219478130340576,
      "learning_rate": 0.0002,
      "loss": 1.2241,
      "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": 324298258513920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}