last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.002290950744558992,
  "eval_steps": 34,
  "global_step": 34,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.738090425173506e-05,
      "eval_loss": 1.0482484102249146,
      "eval_runtime": 1771.3022,
      "eval_samples_per_second": 14.111,
      "eval_steps_per_second": 1.764,
      "step": 1
    },
    {
      "epoch": 0.00020214271275520516,
      "grad_norm": 2.1362860202789307,
      "learning_rate": 1.5e-05,
      "loss": 1.0166,
      "step": 3
    },
    {
      "epoch": 0.0004042854255104103,
      "grad_norm": 1.772956371307373,
      "learning_rate": 3e-05,
      "loss": 1.0004,
      "step": 6
    },
    {
      "epoch": 0.0006064281382656155,
      "grad_norm": 1.3759489059448242,
      "learning_rate": 4.5e-05,
      "loss": 0.9564,
      "step": 9
    },
    {
      "epoch": 0.0008085708510208206,
      "grad_norm": 1.2244105339050293,
      "learning_rate": 4.999675562428437e-05,
      "loss": 0.891,
      "step": 12
    },
    {
      "epoch": 0.001010713563776026,
      "grad_norm": 1.1622825860977173,
      "learning_rate": 4.9979724954289244e-05,
      "loss": 0.8578,
      "step": 15
    },
    {
      "epoch": 0.001212856276531231,
      "grad_norm": 1.089287281036377,
      "learning_rate": 4.994810682835951e-05,
      "loss": 0.832,
      "step": 18
    },
    {
      "epoch": 0.0014149989892864362,
      "grad_norm": 1.685937762260437,
      "learning_rate": 4.990191971059033e-05,
      "loss": 0.8445,
      "step": 21
    },
    {
      "epoch": 0.0016171417020416413,
      "grad_norm": 0.9353536367416382,
      "learning_rate": 4.984119057295783e-05,
      "loss": 0.8481,
      "step": 24
    },
    {
      "epoch": 0.0018192844147968466,
      "grad_norm": 0.9442921876907349,
      "learning_rate": 4.976595487956823e-05,
      "loss": 0.8389,
      "step": 27
    },
    {
      "epoch": 0.002021427127552052,
      "grad_norm": 1.0226161479949951,
      "learning_rate": 4.967625656594782e-05,
      "loss": 0.8224,
      "step": 30
    },
    {
      "epoch": 0.002223569840307257,
      "grad_norm": 0.8120137453079224,
      "learning_rate": 4.957214801338581e-05,
      "loss": 0.849,
      "step": 33
    },
    {
      "epoch": 0.002290950744558992,
      "eval_loss": 0.8384992480278015,
      "eval_runtime": 1781.1941,
      "eval_samples_per_second": 14.033,
      "eval_steps_per_second": 1.754,
      "step": 34
    }
  ],
  "logging_steps": 3,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 34,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.78128332907479e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}