last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0023949227637408694,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00023949227637408692,
      "grad_norm": 5.556643962860107,
      "learning_rate": 2e-05,
      "loss": 2.4864,
      "step": 1
    },
    {
      "epoch": 0.00023949227637408692,
      "eval_loss": 3.196671962738037,
      "eval_runtime": 164.0333,
      "eval_samples_per_second": 5.365,
      "eval_steps_per_second": 5.365,
      "step": 1
    },
    {
      "epoch": 0.00047898455274817384,
      "grad_norm": 4.200049877166748,
      "learning_rate": 4e-05,
      "loss": 3.3358,
      "step": 2
    },
    {
      "epoch": 0.0007184768291222608,
      "grad_norm": 3.4988012313842773,
      "learning_rate": 6e-05,
      "loss": 3.3931,
      "step": 3
    },
    {
      "epoch": 0.0007184768291222608,
      "eval_loss": 3.08847713470459,
      "eval_runtime": 164.5024,
      "eval_samples_per_second": 5.349,
      "eval_steps_per_second": 5.349,
      "step": 3
    },
    {
      "epoch": 0.0009579691054963477,
      "grad_norm": 3.2402377128601074,
      "learning_rate": 8e-05,
      "loss": 2.761,
      "step": 4
    },
    {
      "epoch": 0.0011974613818704347,
      "grad_norm": 3.507751703262329,
      "learning_rate": 0.0001,
      "loss": 2.7889,
      "step": 5
    },
    {
      "epoch": 0.0014369536582445215,
      "grad_norm": 2.906939744949341,
      "learning_rate": 0.00012,
      "loss": 1.8125,
      "step": 6
    },
    {
      "epoch": 0.0014369536582445215,
      "eval_loss": 2.6185643672943115,
      "eval_runtime": 165.0188,
      "eval_samples_per_second": 5.333,
      "eval_steps_per_second": 5.333,
      "step": 6
    },
    {
      "epoch": 0.0016764459346186086,
      "grad_norm": 2.618673324584961,
      "learning_rate": 0.00014,
      "loss": 2.6648,
      "step": 7
    },
    {
      "epoch": 0.0019159382109926954,
      "grad_norm": 1.8121130466461182,
      "learning_rate": 0.00016,
      "loss": 1.9557,
      "step": 8
    },
    {
      "epoch": 0.0021554304873667826,
      "grad_norm": 1.2346168756484985,
      "learning_rate": 0.00018,
      "loss": 1.7235,
      "step": 9
    },
    {
      "epoch": 0.0021554304873667826,
      "eval_loss": 2.1706855297088623,
      "eval_runtime": 163.8793,
      "eval_samples_per_second": 5.37,
      "eval_steps_per_second": 5.37,
      "step": 9
    },
    {
      "epoch": 0.0023949227637408694,
      "grad_norm": 1.9918853044509888,
      "learning_rate": 0.0002,
      "loss": 1.5796,
      "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": 1643164226027520.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}