last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008592910848549946,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008592910848549946,
      "grad_norm": 1.4505956172943115,
      "learning_rate": 2e-05,
      "loss": 2.8565,
      "step": 1
    },
    {
      "epoch": 0.0008592910848549946,
      "eval_loss": 2.9089958667755127,
      "eval_runtime": 15.5121,
      "eval_samples_per_second": 31.588,
      "eval_steps_per_second": 15.794,
      "step": 1
    },
    {
      "epoch": 0.0017185821697099893,
      "grad_norm": 1.3080875873565674,
      "learning_rate": 4e-05,
      "loss": 3.0213,
      "step": 2
    },
    {
      "epoch": 0.0025778732545649837,
      "grad_norm": 2.1423614025115967,
      "learning_rate": 6e-05,
      "loss": 3.052,
      "step": 3
    },
    {
      "epoch": 0.0025778732545649837,
      "eval_loss": 2.899231195449829,
      "eval_runtime": 14.7871,
      "eval_samples_per_second": 33.137,
      "eval_steps_per_second": 16.568,
      "step": 3
    },
    {
      "epoch": 0.0034371643394199786,
      "grad_norm": 1.6318122148513794,
      "learning_rate": 8e-05,
      "loss": 2.9745,
      "step": 4
    },
    {
      "epoch": 0.004296455424274973,
      "grad_norm": 1.2433406114578247,
      "learning_rate": 0.0001,
      "loss": 2.7277,
      "step": 5
    },
    {
      "epoch": 0.0051557465091299675,
      "grad_norm": 1.5990945100784302,
      "learning_rate": 0.00012,
      "loss": 2.6254,
      "step": 6
    },
    {
      "epoch": 0.0051557465091299675,
      "eval_loss": 2.7610018253326416,
      "eval_runtime": 14.8983,
      "eval_samples_per_second": 32.89,
      "eval_steps_per_second": 16.445,
      "step": 6
    },
    {
      "epoch": 0.006015037593984963,
      "grad_norm": 2.0330333709716797,
      "learning_rate": 0.00014,
      "loss": 2.6765,
      "step": 7
    },
    {
      "epoch": 0.006874328678839957,
      "grad_norm": 0.9729666709899902,
      "learning_rate": 0.00016,
      "loss": 2.828,
      "step": 8
    },
    {
      "epoch": 0.007733619763694952,
      "grad_norm": 1.169858694076538,
      "learning_rate": 0.00018,
      "loss": 2.2745,
      "step": 9
    },
    {
      "epoch": 0.007733619763694952,
      "eval_loss": 2.577721357345581,
      "eval_runtime": 14.9257,
      "eval_samples_per_second": 32.829,
      "eval_steps_per_second": 16.415,
      "step": 9
    },
    {
      "epoch": 0.008592910848549946,
      "grad_norm": 1.568127989768982,
      "learning_rate": 0.0002,
      "loss": 2.1253,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}