last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09345794392523364,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009345794392523364,
      "grad_norm": 2.4644651412963867,
      "learning_rate": 2e-05,
      "loss": 1.455,
      "step": 1
    },
    {
      "epoch": 0.009345794392523364,
      "eval_loss": 1.510602355003357,
      "eval_runtime": 16.4984,
      "eval_samples_per_second": 2.728,
      "eval_steps_per_second": 1.394,
      "step": 1
    },
    {
      "epoch": 0.018691588785046728,
      "grad_norm": 2.558793306350708,
      "learning_rate": 4e-05,
      "loss": 1.5614,
      "step": 2
    },
    {
      "epoch": 0.028037383177570093,
      "grad_norm": 2.8438074588775635,
      "learning_rate": 6e-05,
      "loss": 1.6279,
      "step": 3
    },
    {
      "epoch": 0.028037383177570093,
      "eval_loss": 1.4593942165374756,
      "eval_runtime": 16.7044,
      "eval_samples_per_second": 2.694,
      "eval_steps_per_second": 1.377,
      "step": 3
    },
    {
      "epoch": 0.037383177570093455,
      "grad_norm": 3.3598430156707764,
      "learning_rate": 8e-05,
      "loss": 1.7172,
      "step": 4
    },
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 2.6555469036102295,
      "learning_rate": 0.0001,
      "loss": 1.3645,
      "step": 5
    },
    {
      "epoch": 0.056074766355140186,
      "grad_norm": 3.13533616065979,
      "learning_rate": 0.00012,
      "loss": 1.1672,
      "step": 6
    },
    {
      "epoch": 0.056074766355140186,
      "eval_loss": 0.9217208027839661,
      "eval_runtime": 16.8089,
      "eval_samples_per_second": 2.677,
      "eval_steps_per_second": 1.368,
      "step": 6
    },
    {
      "epoch": 0.06542056074766354,
      "grad_norm": 2.7943050861358643,
      "learning_rate": 0.00014,
      "loss": 0.8291,
      "step": 7
    },
    {
      "epoch": 0.07476635514018691,
      "grad_norm": 2.0762441158294678,
      "learning_rate": 0.00016,
      "loss": 0.5093,
      "step": 8
    },
    {
      "epoch": 0.08411214953271028,
      "grad_norm": 1.6523407697677612,
      "learning_rate": 0.00018,
      "loss": 0.2518,
      "step": 9
    },
    {
      "epoch": 0.08411214953271028,
      "eval_loss": 0.1820312887430191,
      "eval_runtime": 16.8474,
      "eval_samples_per_second": 2.671,
      "eval_steps_per_second": 1.365,
      "step": 9
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 1.81255042552948,
      "learning_rate": 0.0002,
      "loss": 0.1513,
      "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": 1.306916403806208e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}