last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003739296264443032,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.739296264443032e-05,
      "grad_norm": 0.5257403254508972,
      "learning_rate": 2e-05,
      "loss": 3.872,
      "step": 1
    },
    {
      "epoch": 3.739296264443032e-05,
      "eval_loss": 0.6931043863296509,
      "eval_runtime": 149.8911,
      "eval_samples_per_second": 75.128,
      "eval_steps_per_second": 37.567,
      "step": 1
    },
    {
      "epoch": 7.478592528886064e-05,
      "grad_norm": 0.5068989396095276,
      "learning_rate": 4e-05,
      "loss": 2.8316,
      "step": 2
    },
    {
      "epoch": 0.00011217888793329095,
      "grad_norm": 0.6643518209457397,
      "learning_rate": 6e-05,
      "loss": 3.1932,
      "step": 3
    },
    {
      "epoch": 0.00011217888793329095,
      "eval_loss": 0.6930645704269409,
      "eval_runtime": 149.8487,
      "eval_samples_per_second": 75.149,
      "eval_steps_per_second": 37.578,
      "step": 3
    },
    {
      "epoch": 0.00014957185057772127,
      "grad_norm": 0.5338044166564941,
      "learning_rate": 8e-05,
      "loss": 2.8836,
      "step": 4
    },
    {
      "epoch": 0.0001869648132221516,
      "grad_norm": 0.6110092997550964,
      "learning_rate": 0.0001,
      "loss": 3.4155,
      "step": 5
    },
    {
      "epoch": 0.0002243577758665819,
      "grad_norm": 0.4845415949821472,
      "learning_rate": 0.00012,
      "loss": 2.6348,
      "step": 6
    },
    {
      "epoch": 0.0002243577758665819,
      "eval_loss": 0.6917085647583008,
      "eval_runtime": 149.9897,
      "eval_samples_per_second": 75.078,
      "eval_steps_per_second": 37.543,
      "step": 6
    },
    {
      "epoch": 0.0002617507385110122,
      "grad_norm": 0.5297170877456665,
      "learning_rate": 0.00014,
      "loss": 2.4877,
      "step": 7
    },
    {
      "epoch": 0.00029914370115544255,
      "grad_norm": 0.7077357769012451,
      "learning_rate": 0.00016,
      "loss": 3.1547,
      "step": 8
    },
    {
      "epoch": 0.0003365366637998729,
      "grad_norm": 0.5892812013626099,
      "learning_rate": 0.00018,
      "loss": 2.7252,
      "step": 9
    },
    {
      "epoch": 0.0003365366637998729,
      "eval_loss": 0.6855435967445374,
      "eval_runtime": 125.6436,
      "eval_samples_per_second": 89.627,
      "eval_steps_per_second": 44.817,
      "step": 9
    },
    {
      "epoch": 0.0003739296264443032,
      "grad_norm": 0.699952244758606,
      "learning_rate": 0.0002,
      "loss": 2.732,
      "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": 23344988553216.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}