last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.022913881993507735,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007637960664502578,
      "eval_loss": NaN,
      "eval_runtime": 20.6863,
      "eval_samples_per_second": 26.684,
      "eval_steps_per_second": 13.342,
      "step": 1
    },
    {
      "epoch": 0.0022913881993507734,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.003818980332251289,
      "eval_loss": NaN,
      "eval_runtime": 20.3317,
      "eval_samples_per_second": 27.15,
      "eval_steps_per_second": 13.575,
      "step": 5
    },
    {
      "epoch": 0.004582776398701547,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00687416459805232,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.007637960664502578,
      "eval_loss": NaN,
      "eval_runtime": 20.3575,
      "eval_samples_per_second": 27.115,
      "eval_steps_per_second": 13.558,
      "step": 10
    },
    {
      "epoch": 0.009165552797403094,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.011456940996753867,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.011456940996753867,
      "eval_loss": NaN,
      "eval_runtime": 20.43,
      "eval_samples_per_second": 27.019,
      "eval_steps_per_second": 13.51,
      "step": 15
    },
    {
      "epoch": 0.01374832919610464,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.015275921329005156,
      "eval_loss": NaN,
      "eval_runtime": 20.3322,
      "eval_samples_per_second": 27.149,
      "eval_steps_per_second": 13.575,
      "step": 20
    },
    {
      "epoch": 0.016039717395455413,
      "grad_norm": NaN,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.018331105594806187,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.019094901661256443,
      "eval_loss": NaN,
      "eval_runtime": 20.3436,
      "eval_samples_per_second": 27.134,
      "eval_steps_per_second": 13.567,
      "step": 25
    },
    {
      "epoch": 0.02062249379415696,
      "grad_norm": NaN,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.022913881993507735,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.022913881993507735,
      "eval_loss": NaN,
      "eval_runtime": 20.3321,
      "eval_samples_per_second": 27.149,
      "eval_steps_per_second": 13.575,
      "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": 2742014856658944.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}