last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.016922859963333804,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005640953321111268,
      "eval_loss": NaN,
      "eval_runtime": 58.3726,
      "eval_samples_per_second": 12.797,
      "eval_steps_per_second": 6.407,
      "step": 1
    },
    {
      "epoch": 0.0016922859963333804,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0028204766605556338,
      "eval_loss": NaN,
      "eval_runtime": 58.4466,
      "eval_samples_per_second": 12.781,
      "eval_steps_per_second": 6.399,
      "step": 5
    },
    {
      "epoch": 0.0033845719926667607,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.005076857989000141,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0056409533211112676,
      "eval_loss": NaN,
      "eval_runtime": 58.6785,
      "eval_samples_per_second": 12.73,
      "eval_steps_per_second": 6.374,
      "step": 10
    },
    {
      "epoch": 0.006769143985333521,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.008461429981666902,
      "grad_norm": NaN,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.008461429981666902,
      "eval_loss": NaN,
      "eval_runtime": 58.6825,
      "eval_samples_per_second": 12.73,
      "eval_steps_per_second": 6.373,
      "step": 15
    },
    {
      "epoch": 0.010153715978000282,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.011281906642222535,
      "eval_loss": NaN,
      "eval_runtime": 58.6995,
      "eval_samples_per_second": 12.726,
      "eval_steps_per_second": 6.371,
      "step": 20
    },
    {
      "epoch": 0.011846001974333662,
      "grad_norm": NaN,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.013538287970667043,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.01410238330277817,
      "eval_loss": NaN,
      "eval_runtime": 58.7649,
      "eval_samples_per_second": 12.712,
      "eval_steps_per_second": 6.364,
      "step": 25
    },
    {
      "epoch": 0.015230573967000423,
      "grad_norm": NaN,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.016922859963333804,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.016922859963333804,
      "eval_loss": NaN,
      "eval_runtime": 58.7804,
      "eval_samples_per_second": 12.708,
      "eval_steps_per_second": 6.363,
      "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": 9801879068344320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}