last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005684240443370754,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005684240443370755,
      "grad_norm": 5.8216376304626465,
      "learning_rate": 2e-05,
      "loss": 10.9298,
      "step": 1
    },
    {
      "epoch": 0.0005684240443370755,
      "eval_loss": 2.1613194942474365,
      "eval_runtime": 11.1079,
      "eval_samples_per_second": 66.709,
      "eval_steps_per_second": 33.4,
      "step": 1
    },
    {
      "epoch": 0.001136848088674151,
      "grad_norm": 4.513489723205566,
      "learning_rate": 4e-05,
      "loss": 6.5667,
      "step": 2
    },
    {
      "epoch": 0.0017052721330112265,
      "grad_norm": 5.058469295501709,
      "learning_rate": 6e-05,
      "loss": 8.6594,
      "step": 3
    },
    {
      "epoch": 0.0017052721330112265,
      "eval_loss": 2.152834415435791,
      "eval_runtime": 11.1059,
      "eval_samples_per_second": 66.722,
      "eval_steps_per_second": 33.406,
      "step": 3
    },
    {
      "epoch": 0.002273696177348302,
      "grad_norm": 5.7656731605529785,
      "learning_rate": 8e-05,
      "loss": 11.6239,
      "step": 4
    },
    {
      "epoch": 0.002842120221685377,
      "grad_norm": 5.779434680938721,
      "learning_rate": 0.0001,
      "loss": 8.2429,
      "step": 5
    },
    {
      "epoch": 0.003410544266022453,
      "grad_norm": 4.3664422035217285,
      "learning_rate": 0.00012,
      "loss": 10.265,
      "step": 6
    },
    {
      "epoch": 0.003410544266022453,
      "eval_loss": 2.0713918209075928,
      "eval_runtime": 11.1833,
      "eval_samples_per_second": 66.26,
      "eval_steps_per_second": 33.175,
      "step": 6
    },
    {
      "epoch": 0.003978968310359528,
      "grad_norm": 4.615260601043701,
      "learning_rate": 0.00014,
      "loss": 9.0162,
      "step": 7
    },
    {
      "epoch": 0.004547392354696604,
      "grad_norm": 4.074900150299072,
      "learning_rate": 0.00016,
      "loss": 7.5965,
      "step": 8
    },
    {
      "epoch": 0.005115816399033679,
      "grad_norm": 3.8367180824279785,
      "learning_rate": 0.00018,
      "loss": 6.7182,
      "step": 9
    },
    {
      "epoch": 0.005115816399033679,
      "eval_loss": 1.9649525880813599,
      "eval_runtime": 11.2009,
      "eval_samples_per_second": 66.155,
      "eval_steps_per_second": 33.122,
      "step": 9
    },
    {
      "epoch": 0.005684240443370754,
      "grad_norm": 3.5268516540527344,
      "learning_rate": 0.0002,
      "loss": 11.8493,
      "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": 84453359616000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}