last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003983799216519488,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00013279330721731626,
      "eval_loss": 10.160923957824707,
      "eval_runtime": 548.4864,
      "eval_samples_per_second": 5.781,
      "eval_steps_per_second": 2.892,
      "step": 1
    },
    {
      "epoch": 0.00039837992165194875,
      "grad_norm": 27.863054275512695,
      "learning_rate": 6e-05,
      "loss": 46.6072,
      "step": 3
    },
    {
      "epoch": 0.0007967598433038975,
      "grad_norm": 35.838321685791016,
      "learning_rate": 0.00012,
      "loss": 31.4981,
      "step": 6
    },
    {
      "epoch": 0.00106234645773853,
      "eval_loss": 7.972387313842773,
      "eval_runtime": 552.0324,
      "eval_samples_per_second": 5.744,
      "eval_steps_per_second": 2.873,
      "step": 8
    },
    {
      "epoch": 0.0011951397649558463,
      "grad_norm": 42.93655014038086,
      "learning_rate": 0.00018,
      "loss": 38.8903,
      "step": 9
    },
    {
      "epoch": 0.001593519686607795,
      "grad_norm": 50.81583023071289,
      "learning_rate": 0.00019510565162951537,
      "loss": 21.0551,
      "step": 12
    },
    {
      "epoch": 0.001991899608259744,
      "grad_norm": 51.02675247192383,
      "learning_rate": 0.00017071067811865476,
      "loss": 20.564,
      "step": 15
    },
    {
      "epoch": 0.00212469291547706,
      "eval_loss": 2.665860652923584,
      "eval_runtime": 552.0831,
      "eval_samples_per_second": 5.744,
      "eval_steps_per_second": 2.873,
      "step": 16
    },
    {
      "epoch": 0.0023902795299116926,
      "grad_norm": 36.232852935791016,
      "learning_rate": 0.00013090169943749476,
      "loss": 7.3799,
      "step": 18
    },
    {
      "epoch": 0.0027886594515636413,
      "grad_norm": 2.943772077560425,
      "learning_rate": 8.435655349597689e-05,
      "loss": 2.4968,
      "step": 21
    },
    {
      "epoch": 0.00318703937321559,
      "grad_norm": 41.44187927246094,
      "learning_rate": 4.12214747707527e-05,
      "loss": 5.0246,
      "step": 24
    },
    {
      "epoch": 0.00318703937321559,
      "eval_loss": 0.9482247233390808,
      "eval_runtime": 552.101,
      "eval_samples_per_second": 5.744,
      "eval_steps_per_second": 2.873,
      "step": 24
    },
    {
      "epoch": 0.0035854192948675387,
      "grad_norm": 20.620649337768555,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 1.3545,
      "step": 27
    },
    {
      "epoch": 0.003983799216519488,
      "grad_norm": 26.846813201904297,
      "learning_rate": 0.0,
      "loss": 2.5073,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 70,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5391528134377472e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}