last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009643201542912247,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003214400514304082,
      "eval_loss": 2.958423137664795,
      "eval_runtime": 198.7725,
      "eval_samples_per_second": 6.59,
      "eval_steps_per_second": 3.295,
      "step": 1
    },
    {
      "epoch": 0.0009643201542912247,
      "grad_norm": 4.432087421417236,
      "learning_rate": 3e-05,
      "loss": 11.2323,
      "step": 3
    },
    {
      "epoch": 0.0019286403085824494,
      "grad_norm": 4.153701305389404,
      "learning_rate": 6e-05,
      "loss": 12.1651,
      "step": 6
    },
    {
      "epoch": 0.0025715204114432656,
      "eval_loss": 2.7854690551757812,
      "eval_runtime": 200.1028,
      "eval_samples_per_second": 6.547,
      "eval_steps_per_second": 3.273,
      "step": 8
    },
    {
      "epoch": 0.0028929604628736743,
      "grad_norm": 4.62204647064209,
      "learning_rate": 9e-05,
      "loss": 11.5557,
      "step": 9
    },
    {
      "epoch": 0.003857280617164899,
      "grad_norm": 3.8450965881347656,
      "learning_rate": 9.755282581475769e-05,
      "loss": 10.8073,
      "step": 12
    },
    {
      "epoch": 0.0048216007714561235,
      "grad_norm": 4.419191360473633,
      "learning_rate": 8.535533905932738e-05,
      "loss": 9.5406,
      "step": 15
    },
    {
      "epoch": 0.005143040822886531,
      "eval_loss": 2.2781641483306885,
      "eval_runtime": 200.068,
      "eval_samples_per_second": 6.548,
      "eval_steps_per_second": 3.274,
      "step": 16
    },
    {
      "epoch": 0.0057859209257473485,
      "grad_norm": 4.035886764526367,
      "learning_rate": 6.545084971874738e-05,
      "loss": 9.4144,
      "step": 18
    },
    {
      "epoch": 0.006750241080038573,
      "grad_norm": 3.005903482437134,
      "learning_rate": 4.2178276747988446e-05,
      "loss": 8.6917,
      "step": 21
    },
    {
      "epoch": 0.007714561234329798,
      "grad_norm": 3.32614803314209,
      "learning_rate": 2.061073738537635e-05,
      "loss": 8.2647,
      "step": 24
    },
    {
      "epoch": 0.007714561234329798,
      "eval_loss": 2.131072998046875,
      "eval_runtime": 200.0895,
      "eval_samples_per_second": 6.547,
      "eval_steps_per_second": 3.274,
      "step": 24
    },
    {
      "epoch": 0.008678881388621022,
      "grad_norm": 2.954596996307373,
      "learning_rate": 5.449673790581611e-06,
      "loss": 8.3838,
      "step": 27
    },
    {
      "epoch": 0.009643201542912247,
      "grad_norm": 2.7677104473114014,
      "learning_rate": 0.0,
      "loss": 8.3747,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0183366485504e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}