last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.032706459525756335,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003270645952575634,
      "grad_norm": 1.391844391822815,
      "learning_rate": 2e-05,
      "loss": 8.3848,
      "step": 1
    },
    {
      "epoch": 0.003270645952575634,
      "eval_loss": 2.297170400619507,
      "eval_runtime": 4.4406,
      "eval_samples_per_second": 29.05,
      "eval_steps_per_second": 14.638,
      "step": 1
    },
    {
      "epoch": 0.006541291905151268,
      "grad_norm": 1.952135443687439,
      "learning_rate": 4e-05,
      "loss": 10.5925,
      "step": 2
    },
    {
      "epoch": 0.009811937857726901,
      "grad_norm": 1.852607250213623,
      "learning_rate": 6e-05,
      "loss": 11.9831,
      "step": 3
    },
    {
      "epoch": 0.009811937857726901,
      "eval_loss": 2.2972500324249268,
      "eval_runtime": 3.4645,
      "eval_samples_per_second": 37.235,
      "eval_steps_per_second": 18.762,
      "step": 3
    },
    {
      "epoch": 0.013082583810302535,
      "grad_norm": 1.4563934803009033,
      "learning_rate": 8e-05,
      "loss": 7.0282,
      "step": 4
    },
    {
      "epoch": 0.016353229762878167,
      "grad_norm": 3.0670440196990967,
      "learning_rate": 0.0001,
      "loss": 9.6027,
      "step": 5
    },
    {
      "epoch": 0.019623875715453803,
      "grad_norm": 1.7290756702423096,
      "learning_rate": 0.00012,
      "loss": 8.393,
      "step": 6
    },
    {
      "epoch": 0.019623875715453803,
      "eval_loss": 2.274766445159912,
      "eval_runtime": 3.5282,
      "eval_samples_per_second": 36.563,
      "eval_steps_per_second": 18.423,
      "step": 6
    },
    {
      "epoch": 0.022894521668029435,
      "grad_norm": 2.7738356590270996,
      "learning_rate": 0.00014,
      "loss": 9.1987,
      "step": 7
    },
    {
      "epoch": 0.02616516762060507,
      "grad_norm": 4.28538179397583,
      "learning_rate": 0.00016,
      "loss": 12.1198,
      "step": 8
    },
    {
      "epoch": 0.029435813573180702,
      "grad_norm": 4.097850799560547,
      "learning_rate": 0.00018,
      "loss": 8.7893,
      "step": 9
    },
    {
      "epoch": 0.029435813573180702,
      "eval_loss": 2.1066341400146484,
      "eval_runtime": 3.6131,
      "eval_samples_per_second": 35.703,
      "eval_steps_per_second": 17.99,
      "step": 9
    },
    {
      "epoch": 0.032706459525756335,
      "grad_norm": 4.777021884918213,
      "learning_rate": 0.0002,
      "loss": 7.9144,
      "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": 1632478030725120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}