last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0039012971813127866,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00039012971813127864,
      "grad_norm": 3.517857074737549,
      "learning_rate": 2e-05,
      "loss": 2.9308,
      "step": 1
    },
    {
      "epoch": 0.00039012971813127864,
      "eval_loss": 0.7795681357383728,
      "eval_runtime": 57.8942,
      "eval_samples_per_second": 18.655,
      "eval_steps_per_second": 9.327,
      "step": 1
    },
    {
      "epoch": 0.0007802594362625573,
      "grad_norm": 4.007322311401367,
      "learning_rate": 4e-05,
      "loss": 2.9124,
      "step": 2
    },
    {
      "epoch": 0.001170389154393836,
      "grad_norm": 3.6519687175750732,
      "learning_rate": 6e-05,
      "loss": 2.921,
      "step": 3
    },
    {
      "epoch": 0.001170389154393836,
      "eval_loss": 0.7768089175224304,
      "eval_runtime": 57.8166,
      "eval_samples_per_second": 18.68,
      "eval_steps_per_second": 9.34,
      "step": 3
    },
    {
      "epoch": 0.0015605188725251145,
      "grad_norm": 3.2878363132476807,
      "learning_rate": 8e-05,
      "loss": 2.84,
      "step": 4
    },
    {
      "epoch": 0.0019506485906563933,
      "grad_norm": 4.44495964050293,
      "learning_rate": 0.0001,
      "loss": 3.7476,
      "step": 5
    },
    {
      "epoch": 0.002340778308787672,
      "grad_norm": 3.2834181785583496,
      "learning_rate": 0.00012,
      "loss": 2.8451,
      "step": 6
    },
    {
      "epoch": 0.002340778308787672,
      "eval_loss": 0.7554607391357422,
      "eval_runtime": 57.8765,
      "eval_samples_per_second": 18.66,
      "eval_steps_per_second": 9.33,
      "step": 6
    },
    {
      "epoch": 0.0027309080269189503,
      "grad_norm": 3.4334750175476074,
      "learning_rate": 0.00014,
      "loss": 3.3178,
      "step": 7
    },
    {
      "epoch": 0.003121037745050229,
      "grad_norm": 3.7431459426879883,
      "learning_rate": 0.00016,
      "loss": 3.3982,
      "step": 8
    },
    {
      "epoch": 0.003511167463181508,
      "grad_norm": 3.4571352005004883,
      "learning_rate": 0.00018,
      "loss": 3.3138,
      "step": 9
    },
    {
      "epoch": 0.003511167463181508,
      "eval_loss": 0.7286794185638428,
      "eval_runtime": 57.5525,
      "eval_samples_per_second": 18.765,
      "eval_steps_per_second": 9.383,
      "step": 9
    },
    {
      "epoch": 0.0039012971813127866,
      "grad_norm": 3.0988426208496094,
      "learning_rate": 0.0002,
      "loss": 3.2137,
      "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": 3461522827247616.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}