last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0012221299531516852,
  "eval_steps": 5,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.1336665465427825e-05,
      "eval_loss": 11.820560455322266,
      "eval_runtime": 130.7138,
      "eval_samples_per_second": 102.797,
      "eval_steps_per_second": 51.402,
      "step": 1
    },
    {
      "epoch": 9.400999639628348e-05,
      "grad_norm": 9.347860336303711,
      "learning_rate": 2.857142857142857e-05,
      "loss": 11.7883,
      "step": 3
    },
    {
      "epoch": 0.00015668332732713913,
      "eval_loss": 11.753523826599121,
      "eval_runtime": 130.2236,
      "eval_samples_per_second": 103.184,
      "eval_steps_per_second": 51.596,
      "step": 5
    },
    {
      "epoch": 0.00018801999279256695,
      "grad_norm": 8.440356254577637,
      "learning_rate": 5.714285714285714e-05,
      "loss": 11.7292,
      "step": 6
    },
    {
      "epoch": 0.0002820299891888504,
      "grad_norm": 7.404284954071045,
      "learning_rate": 8.571428571428571e-05,
      "loss": 11.5233,
      "step": 9
    },
    {
      "epoch": 0.00031336665465427825,
      "eval_loss": 11.403406143188477,
      "eval_runtime": 130.469,
      "eval_samples_per_second": 102.99,
      "eval_steps_per_second": 51.499,
      "step": 10
    },
    {
      "epoch": 0.0003760399855851339,
      "grad_norm": 7.426136493682861,
      "learning_rate": 0.00011428571428571428,
      "loss": 11.4367,
      "step": 12
    },
    {
      "epoch": 0.00047004998198141735,
      "grad_norm": 7.491276741027832,
      "learning_rate": 0.00014285714285714287,
      "loss": 11.28,
      "step": 15
    },
    {
      "epoch": 0.00047004998198141735,
      "eval_loss": 10.960875511169434,
      "eval_runtime": 129.9537,
      "eval_samples_per_second": 103.398,
      "eval_steps_per_second": 51.703,
      "step": 15
    },
    {
      "epoch": 0.0005640599783777008,
      "grad_norm": 6.318249225616455,
      "learning_rate": 0.00017142857142857143,
      "loss": 10.8666,
      "step": 18
    },
    {
      "epoch": 0.0006267333093085565,
      "eval_loss": 10.430113792419434,
      "eval_runtime": 129.8947,
      "eval_samples_per_second": 103.445,
      "eval_steps_per_second": 51.727,
      "step": 20
    },
    {
      "epoch": 0.0006580699747739844,
      "grad_norm": 5.295856475830078,
      "learning_rate": 0.0002,
      "loss": 10.565,
      "step": 21
    },
    {
      "epoch": 0.0007520799711702678,
      "grad_norm": 4.21265172958374,
      "learning_rate": 0.00018660254037844388,
      "loss": 10.1523,
      "step": 24
    },
    {
      "epoch": 0.0007834166366356955,
      "eval_loss": 9.9229154586792,
      "eval_runtime": 130.5123,
      "eval_samples_per_second": 102.956,
      "eval_steps_per_second": 51.482,
      "step": 25
    },
    {
      "epoch": 0.0008460899675665512,
      "grad_norm": 4.139631271362305,
      "learning_rate": 0.00015000000000000001,
      "loss": 9.9544,
      "step": 27
    },
    {
      "epoch": 0.0009400999639628347,
      "grad_norm": 3.844069242477417,
      "learning_rate": 0.0001,
      "loss": 9.713,
      "step": 30
    },
    {
      "epoch": 0.0009400999639628347,
      "eval_loss": 9.596590995788574,
      "eval_runtime": 130.3595,
      "eval_samples_per_second": 103.076,
      "eval_steps_per_second": 51.542,
      "step": 30
    },
    {
      "epoch": 0.0010341099603591183,
      "grad_norm": 3.5350520610809326,
      "learning_rate": 5.000000000000002e-05,
      "loss": 9.5719,
      "step": 33
    },
    {
      "epoch": 0.0010967832912899737,
      "eval_loss": 9.454024314880371,
      "eval_runtime": 130.2285,
      "eval_samples_per_second": 103.18,
      "eval_steps_per_second": 51.594,
      "step": 35
    },
    {
      "epoch": 0.0011281199567554016,
      "grad_norm": 2.883392095565796,
      "learning_rate": 1.339745962155613e-05,
      "loss": 9.4912,
      "step": 36
    },
    {
      "epoch": 0.0012221299531516852,
      "grad_norm": 2.996534824371338,
      "learning_rate": 0.0,
      "loss": 9.4766,
      "step": 39
    }
  ],
  "logging_steps": 3,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 21,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5294156414976.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}