last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.025789813023855575,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025789813023855577,
      "grad_norm": 17.448917388916016,
      "learning_rate": 2e-05,
      "loss": 8.6638,
      "step": 1
    },
    {
      "epoch": 0.0025789813023855577,
      "eval_loss": 9.067191123962402,
      "eval_runtime": 25.4176,
      "eval_samples_per_second": 6.452,
      "eval_steps_per_second": 3.226,
      "step": 1
    },
    {
      "epoch": 0.0051579626047711154,
      "grad_norm": 16.798805236816406,
      "learning_rate": 4e-05,
      "loss": 9.4982,
      "step": 2
    },
    {
      "epoch": 0.007736943907156673,
      "grad_norm": 16.873533248901367,
      "learning_rate": 6e-05,
      "loss": 9.3925,
      "step": 3
    },
    {
      "epoch": 0.007736943907156673,
      "eval_loss": 8.867646217346191,
      "eval_runtime": 25.3746,
      "eval_samples_per_second": 6.463,
      "eval_steps_per_second": 3.232,
      "step": 3
    },
    {
      "epoch": 0.010315925209542231,
      "grad_norm": 18.353174209594727,
      "learning_rate": 8e-05,
      "loss": 9.1951,
      "step": 4
    },
    {
      "epoch": 0.012894906511927788,
      "grad_norm": 15.738860130310059,
      "learning_rate": 0.0001,
      "loss": 8.0012,
      "step": 5
    },
    {
      "epoch": 0.015473887814313346,
      "grad_norm": 16.180191040039062,
      "learning_rate": 0.00012,
      "loss": 7.3197,
      "step": 6
    },
    {
      "epoch": 0.015473887814313346,
      "eval_loss": 6.428730487823486,
      "eval_runtime": 23.797,
      "eval_samples_per_second": 6.892,
      "eval_steps_per_second": 3.446,
      "step": 6
    },
    {
      "epoch": 0.018052869116698903,
      "grad_norm": 12.541686058044434,
      "learning_rate": 0.00014,
      "loss": 5.5577,
      "step": 7
    },
    {
      "epoch": 0.020631850419084462,
      "grad_norm": 11.04792308807373,
      "learning_rate": 0.00016,
      "loss": 5.01,
      "step": 8
    },
    {
      "epoch": 0.02321083172147002,
      "grad_norm": 12.31761646270752,
      "learning_rate": 0.00018,
      "loss": 5.0647,
      "step": 9
    },
    {
      "epoch": 0.02321083172147002,
      "eval_loss": 3.7855212688446045,
      "eval_runtime": 23.7876,
      "eval_samples_per_second": 6.894,
      "eval_steps_per_second": 3.447,
      "step": 9
    },
    {
      "epoch": 0.025789813023855575,
      "grad_norm": 16.142803192138672,
      "learning_rate": 0.0002,
      "loss": 4.0176,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1935445139128320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}