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": 5.190128803253174,
      "learning_rate": 2e-05,
      "loss": 2.4864,
      "step": 1
    },
    {
      "epoch": 0.00023949227637408692,
      "eval_loss": 3.196671962738037,
      "eval_runtime": 163.6418,
      "eval_samples_per_second": 5.378,
      "eval_steps_per_second": 5.378,
      "step": 1
    },
    {
      "epoch": 0.00047898455274817384,
      "grad_norm": 3.8827998638153076,
      "learning_rate": 4e-05,
      "loss": 3.3358,
      "step": 2
    },
    {
      "epoch": 0.0007184768291222608,
      "grad_norm": 3.203981876373291,
      "learning_rate": 6e-05,
      "loss": 3.3946,
      "step": 3
    },
    {
      "epoch": 0.0007184768291222608,
      "eval_loss": 3.0946106910705566,
      "eval_runtime": 162.7901,
      "eval_samples_per_second": 5.406,
      "eval_steps_per_second": 5.406,
      "step": 3
    },
    {
      "epoch": 0.0009579691054963477,
      "grad_norm": 3.0879592895507812,
      "learning_rate": 8e-05,
      "loss": 2.7666,
      "step": 4
    },
    {
      "epoch": 0.0011974613818704347,
      "grad_norm": 3.3717029094696045,
      "learning_rate": 0.0001,
      "loss": 2.7991,
      "step": 5
    },
    {
      "epoch": 0.0014369536582445215,
      "grad_norm": 2.83575439453125,
      "learning_rate": 0.00012,
      "loss": 1.8184,
      "step": 6
    },
    {
      "epoch": 0.0014369536582445215,
      "eval_loss": 2.6198861598968506,
      "eval_runtime": 162.8545,
      "eval_samples_per_second": 5.404,
      "eval_steps_per_second": 5.404,
      "step": 6
    },
    {
      "epoch": 0.0016764459346186086,
      "grad_norm": 2.426558256149292,
      "learning_rate": 0.00014,
      "loss": 2.6655,
      "step": 7
    },
    {
      "epoch": 0.0019159382109926954,
      "grad_norm": 1.7445547580718994,
      "learning_rate": 0.00016,
      "loss": 1.9561,
      "step": 8
    },
    {
      "epoch": 0.0021554304873667826,
      "grad_norm": 1.1460472345352173,
      "learning_rate": 0.00018,
      "loss": 1.7204,
      "step": 9
    },
    {
      "epoch": 0.0021554304873667826,
      "eval_loss": 2.171725034713745,
      "eval_runtime": 163.204,
      "eval_samples_per_second": 5.392,
      "eval_steps_per_second": 5.392,
      "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
}