last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00914703864623828,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000914703864623828,
      "grad_norm": 1.6017396450042725,
      "learning_rate": 2e-05,
      "loss": 2.1049,
      "step": 1
    },
    {
      "epoch": 0.000914703864623828,
      "eval_loss": 1.9489076137542725,
      "eval_runtime": 15.793,
      "eval_samples_per_second": 29.19,
      "eval_steps_per_second": 14.627,
      "step": 1
    },
    {
      "epoch": 0.001829407729247656,
      "grad_norm": 1.5895532369613647,
      "learning_rate": 4e-05,
      "loss": 1.9172,
      "step": 2
    },
    {
      "epoch": 0.002744111593871484,
      "grad_norm": 1.5415589809417725,
      "learning_rate": 6e-05,
      "loss": 1.9068,
      "step": 3
    },
    {
      "epoch": 0.002744111593871484,
      "eval_loss": 1.923887014389038,
      "eval_runtime": 15.7936,
      "eval_samples_per_second": 29.189,
      "eval_steps_per_second": 14.626,
      "step": 3
    },
    {
      "epoch": 0.003658815458495312,
      "grad_norm": 1.4512304067611694,
      "learning_rate": 8e-05,
      "loss": 1.8857,
      "step": 4
    },
    {
      "epoch": 0.00457351932311914,
      "grad_norm": 1.7237379550933838,
      "learning_rate": 0.0001,
      "loss": 1.8861,
      "step": 5
    },
    {
      "epoch": 0.005488223187742968,
      "grad_norm": 1.8233637809753418,
      "learning_rate": 0.00012,
      "loss": 1.8641,
      "step": 6
    },
    {
      "epoch": 0.005488223187742968,
      "eval_loss": 1.602256417274475,
      "eval_runtime": 15.834,
      "eval_samples_per_second": 29.115,
      "eval_steps_per_second": 14.589,
      "step": 6
    },
    {
      "epoch": 0.006402927052366796,
      "grad_norm": 1.6963632106781006,
      "learning_rate": 0.00014,
      "loss": 1.5499,
      "step": 7
    },
    {
      "epoch": 0.007317630916990624,
      "grad_norm": 1.6420607566833496,
      "learning_rate": 0.00016,
      "loss": 1.4424,
      "step": 8
    },
    {
      "epoch": 0.008232334781614453,
      "grad_norm": 1.7176631689071655,
      "learning_rate": 0.00018,
      "loss": 1.3188,
      "step": 9
    },
    {
      "epoch": 0.008232334781614453,
      "eval_loss": 1.27040696144104,
      "eval_runtime": 15.8341,
      "eval_samples_per_second": 29.114,
      "eval_steps_per_second": 14.589,
      "step": 9
    },
    {
      "epoch": 0.00914703864623828,
      "grad_norm": 2.091480016708374,
      "learning_rate": 0.0002,
      "loss": 1.1523,
      "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": 1911091416268800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}