last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.023837902264600714,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0023837902264600714,
      "grad_norm": 1.388173222541809,
      "learning_rate": 2e-05,
      "loss": 3.4581,
      "step": 1
    },
    {
      "epoch": 0.0023837902264600714,
      "eval_loss": 4.8837971687316895,
      "eval_runtime": 23.6073,
      "eval_samples_per_second": 7.498,
      "eval_steps_per_second": 3.77,
      "step": 1
    },
    {
      "epoch": 0.004767580452920143,
      "grad_norm": 1.1888856887817383,
      "learning_rate": 4e-05,
      "loss": 2.9179,
      "step": 2
    },
    {
      "epoch": 0.007151370679380214,
      "grad_norm": 1.1461793184280396,
      "learning_rate": 6e-05,
      "loss": 2.6894,
      "step": 3
    },
    {
      "epoch": 0.007151370679380214,
      "eval_loss": 4.878833293914795,
      "eval_runtime": 23.0772,
      "eval_samples_per_second": 7.67,
      "eval_steps_per_second": 3.857,
      "step": 3
    },
    {
      "epoch": 0.009535160905840286,
      "grad_norm": 1.5066274404525757,
      "learning_rate": 8e-05,
      "loss": 3.0156,
      "step": 4
    },
    {
      "epoch": 0.011918951132300357,
      "grad_norm": 1.1865642070770264,
      "learning_rate": 0.0001,
      "loss": 2.993,
      "step": 5
    },
    {
      "epoch": 0.014302741358760428,
      "grad_norm": 1.3507497310638428,
      "learning_rate": 0.00012,
      "loss": 2.5632,
      "step": 6
    },
    {
      "epoch": 0.014302741358760428,
      "eval_loss": 4.79640007019043,
      "eval_runtime": 23.1984,
      "eval_samples_per_second": 7.63,
      "eval_steps_per_second": 3.836,
      "step": 6
    },
    {
      "epoch": 0.0166865315852205,
      "grad_norm": 1.5210137367248535,
      "learning_rate": 0.00014,
      "loss": 3.0109,
      "step": 7
    },
    {
      "epoch": 0.01907032181168057,
      "grad_norm": 1.6760146617889404,
      "learning_rate": 0.00016,
      "loss": 2.6024,
      "step": 8
    },
    {
      "epoch": 0.021454112038140644,
      "grad_norm": 2.1592094898223877,
      "learning_rate": 0.00018,
      "loss": 2.3319,
      "step": 9
    },
    {
      "epoch": 0.021454112038140644,
      "eval_loss": 4.315332889556885,
      "eval_runtime": 23.146,
      "eval_samples_per_second": 7.647,
      "eval_steps_per_second": 3.845,
      "step": 9
    },
    {
      "epoch": 0.023837902264600714,
      "grad_norm": 1.8350367546081543,
      "learning_rate": 0.0002,
      "loss": 2.9011,
      "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": 3699128496291840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}