last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0021554304873667826,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00023949227637408692,
      "grad_norm": 5.223796367645264,
      "learning_rate": 2e-05,
      "loss": 2.4864,
      "step": 1
    },
    {
      "epoch": 0.00023949227637408692,
      "eval_loss": 3.196671962738037,
      "eval_runtime": 163.78,
      "eval_samples_per_second": 5.373,
      "eval_steps_per_second": 5.373,
      "step": 1
    },
    {
      "epoch": 0.00047898455274817384,
      "grad_norm": 3.92407488822937,
      "learning_rate": 4e-05,
      "loss": 3.3358,
      "step": 2
    },
    {
      "epoch": 0.0007184768291222608,
      "grad_norm": 3.1619839668273926,
      "learning_rate": 6e-05,
      "loss": 3.3959,
      "step": 3
    },
    {
      "epoch": 0.0007184768291222608,
      "eval_loss": 3.0963945388793945,
      "eval_runtime": 163.8871,
      "eval_samples_per_second": 5.37,
      "eval_steps_per_second": 5.37,
      "step": 3
    },
    {
      "epoch": 0.0009579691054963477,
      "grad_norm": 3.1326329708099365,
      "learning_rate": 8e-05,
      "loss": 2.768,
      "step": 4
    },
    {
      "epoch": 0.0011974613818704347,
      "grad_norm": 3.3326337337493896,
      "learning_rate": 0.0001,
      "loss": 2.8028,
      "step": 5
    },
    {
      "epoch": 0.0014369536582445215,
      "grad_norm": 2.820315361022949,
      "learning_rate": 0.00012,
      "loss": 1.8255,
      "step": 6
    },
    {
      "epoch": 0.0014369536582445215,
      "eval_loss": 2.628291368484497,
      "eval_runtime": 163.4655,
      "eval_samples_per_second": 5.383,
      "eval_steps_per_second": 5.383,
      "step": 6
    },
    {
      "epoch": 0.0016764459346186086,
      "grad_norm": 2.3316752910614014,
      "learning_rate": 0.00014,
      "loss": 2.6749,
      "step": 7
    },
    {
      "epoch": 0.0019159382109926954,
      "grad_norm": 1.7211425304412842,
      "learning_rate": 0.00016,
      "loss": 1.9641,
      "step": 8
    },
    {
      "epoch": 0.0021554304873667826,
      "grad_norm": 1.1501375436782837,
      "learning_rate": 0.00018,
      "loss": 1.7288,
      "step": 9
    },
    {
      "epoch": 0.0021554304873667826,
      "eval_loss": 2.1707475185394287,
      "eval_runtime": 163.4934,
      "eval_samples_per_second": 5.382,
      "eval_steps_per_second": 5.382,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1478847803424768.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}