last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005471207769115032,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005471207769115032,
      "grad_norm": 60.101470947265625,
      "learning_rate": 2e-05,
      "loss": 14.7588,
      "step": 1
    },
    {
      "epoch": 0.0005471207769115032,
      "eval_loss": 4.2187299728393555,
      "eval_runtime": 55.2355,
      "eval_samples_per_second": 13.94,
      "eval_steps_per_second": 6.97,
      "step": 1
    },
    {
      "epoch": 0.0010942415538230065,
      "grad_norm": 57.0677375793457,
      "learning_rate": 4e-05,
      "loss": 18.6801,
      "step": 2
    },
    {
      "epoch": 0.0016413623307345096,
      "grad_norm": 61.981815338134766,
      "learning_rate": 6e-05,
      "loss": 18.8408,
      "step": 3
    },
    {
      "epoch": 0.0016413623307345096,
      "eval_loss": 4.150631904602051,
      "eval_runtime": 52.0247,
      "eval_samples_per_second": 14.801,
      "eval_steps_per_second": 7.4,
      "step": 3
    },
    {
      "epoch": 0.002188483107646013,
      "grad_norm": 45.864688873291016,
      "learning_rate": 8e-05,
      "loss": 13.6681,
      "step": 4
    },
    {
      "epoch": 0.002735603884557516,
      "grad_norm": 69.2713851928711,
      "learning_rate": 0.0001,
      "loss": 15.6618,
      "step": 5
    },
    {
      "epoch": 0.003282724661469019,
      "grad_norm": 48.16803741455078,
      "learning_rate": 0.00012,
      "loss": 13.5545,
      "step": 6
    },
    {
      "epoch": 0.003282724661469019,
      "eval_loss": 3.492276906967163,
      "eval_runtime": 53.1738,
      "eval_samples_per_second": 14.481,
      "eval_steps_per_second": 7.24,
      "step": 6
    },
    {
      "epoch": 0.0038298454383805225,
      "grad_norm": 42.976959228515625,
      "learning_rate": 0.00014,
      "loss": 13.2728,
      "step": 7
    },
    {
      "epoch": 0.004376966215292026,
      "grad_norm": 45.73899459838867,
      "learning_rate": 0.00016,
      "loss": 12.5971,
      "step": 8
    },
    {
      "epoch": 0.004924086992203529,
      "grad_norm": 51.917232513427734,
      "learning_rate": 0.00018,
      "loss": 13.5721,
      "step": 9
    },
    {
      "epoch": 0.004924086992203529,
      "eval_loss": 2.8099541664123535,
      "eval_runtime": 53.7991,
      "eval_samples_per_second": 14.313,
      "eval_steps_per_second": 7.156,
      "step": 9
    },
    {
      "epoch": 0.005471207769115032,
      "grad_norm": 59.5560417175293,
      "learning_rate": 0.0002,
      "loss": 10.3366,
      "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": 1796487562395648.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}