last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008618275052248292,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.618275052248293e-05,
      "grad_norm": 1.4272311925888062,
      "learning_rate": 2e-05,
      "loss": 1.7574,
      "step": 1
    },
    {
      "epoch": 8.618275052248293e-05,
      "eval_loss": NaN,
      "eval_runtime": 196.86,
      "eval_samples_per_second": 24.82,
      "eval_steps_per_second": 12.41,
      "step": 1
    },
    {
      "epoch": 0.00017236550104496586,
      "grad_norm": 1.0502504110336304,
      "learning_rate": 4e-05,
      "loss": 1.6179,
      "step": 2
    },
    {
      "epoch": 0.0002585482515674488,
      "grad_norm": 1.1126679182052612,
      "learning_rate": 6e-05,
      "loss": 1.3261,
      "step": 3
    },
    {
      "epoch": 0.0002585482515674488,
      "eval_loss": NaN,
      "eval_runtime": 196.7855,
      "eval_samples_per_second": 24.829,
      "eval_steps_per_second": 12.415,
      "step": 3
    },
    {
      "epoch": 0.0003447310020899317,
      "grad_norm": 1.6080414056777954,
      "learning_rate": 8e-05,
      "loss": 1.7585,
      "step": 4
    },
    {
      "epoch": 0.0004309137526124146,
      "grad_norm": 1.2546119689941406,
      "learning_rate": 0.0001,
      "loss": 1.4873,
      "step": 5
    },
    {
      "epoch": 0.0005170965031348976,
      "grad_norm": 1.3804237842559814,
      "learning_rate": 0.00012,
      "loss": 1.7858,
      "step": 6
    },
    {
      "epoch": 0.0005170965031348976,
      "eval_loss": NaN,
      "eval_runtime": 196.9055,
      "eval_samples_per_second": 24.814,
      "eval_steps_per_second": 12.407,
      "step": 6
    },
    {
      "epoch": 0.0006032792536573804,
      "grad_norm": 1.0448002815246582,
      "learning_rate": 0.00014,
      "loss": 1.0582,
      "step": 7
    },
    {
      "epoch": 0.0006894620041798634,
      "grad_norm": 1.4085164070129395,
      "learning_rate": 0.00016,
      "loss": 1.7012,
      "step": 8
    },
    {
      "epoch": 0.0007756447547023463,
      "grad_norm": 2.224423885345459,
      "learning_rate": 0.00018,
      "loss": 2.6606,
      "step": 9
    },
    {
      "epoch": 0.0007756447547023463,
      "eval_loss": NaN,
      "eval_runtime": 196.8308,
      "eval_samples_per_second": 24.823,
      "eval_steps_per_second": 12.412,
      "step": 9
    },
    {
      "epoch": 0.0008618275052248292,
      "grad_norm": 0.6651121973991394,
      "learning_rate": 0.0002,
      "loss": 0.6408,
      "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": 1189070564229120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}