last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00211999152003392,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00021199915200339198,
      "grad_norm": 0.9828506708145142,
      "learning_rate": 2e-05,
      "loss": 2.2407,
      "step": 1
    },
    {
      "epoch": 0.00021199915200339198,
      "eval_loss": 2.1086606979370117,
      "eval_runtime": 76.5017,
      "eval_samples_per_second": 25.973,
      "eval_steps_per_second": 12.993,
      "step": 1
    },
    {
      "epoch": 0.00042399830400678397,
      "grad_norm": 1.0008634328842163,
      "learning_rate": 4e-05,
      "loss": 2.3608,
      "step": 2
    },
    {
      "epoch": 0.0006359974560101759,
      "grad_norm": 1.0057114362716675,
      "learning_rate": 6e-05,
      "loss": 2.112,
      "step": 3
    },
    {
      "epoch": 0.0006359974560101759,
      "eval_loss": 2.096790075302124,
      "eval_runtime": 76.717,
      "eval_samples_per_second": 25.9,
      "eval_steps_per_second": 12.957,
      "step": 3
    },
    {
      "epoch": 0.0008479966080135679,
      "grad_norm": 1.0478014945983887,
      "learning_rate": 8e-05,
      "loss": 1.9836,
      "step": 4
    },
    {
      "epoch": 0.00105999576001696,
      "grad_norm": 1.0004569292068481,
      "learning_rate": 0.0001,
      "loss": 2.211,
      "step": 5
    },
    {
      "epoch": 0.0012719949120203519,
      "grad_norm": 0.8505454063415527,
      "learning_rate": 0.00012,
      "loss": 1.9588,
      "step": 6
    },
    {
      "epoch": 0.0012719949120203519,
      "eval_loss": 1.9562186002731323,
      "eval_runtime": 76.5847,
      "eval_samples_per_second": 25.945,
      "eval_steps_per_second": 12.979,
      "step": 6
    },
    {
      "epoch": 0.001483994064023744,
      "grad_norm": 0.7165520191192627,
      "learning_rate": 0.00014,
      "loss": 1.8718,
      "step": 7
    },
    {
      "epoch": 0.0016959932160271359,
      "grad_norm": 0.75093674659729,
      "learning_rate": 0.00016,
      "loss": 1.5151,
      "step": 8
    },
    {
      "epoch": 0.0019079923680305278,
      "grad_norm": 0.7422349452972412,
      "learning_rate": 0.00018,
      "loss": 2.0939,
      "step": 9
    },
    {
      "epoch": 0.0019079923680305278,
      "eval_loss": 1.8474469184875488,
      "eval_runtime": 76.6376,
      "eval_samples_per_second": 25.927,
      "eval_steps_per_second": 12.97,
      "step": 9
    },
    {
      "epoch": 0.00211999152003392,
      "grad_norm": 0.8010254502296448,
      "learning_rate": 0.0002,
      "loss": 1.9063,
      "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": 1000194661416960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}