last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0032144005143040825,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003214400514304082,
      "grad_norm": 2.314427375793457,
      "learning_rate": 2e-05,
      "loss": 8.2829,
      "step": 1
    },
    {
      "epoch": 0.0003214400514304082,
      "eval_loss": 2.0177955627441406,
      "eval_runtime": 60.7423,
      "eval_samples_per_second": 21.567,
      "eval_steps_per_second": 10.783,
      "step": 1
    },
    {
      "epoch": 0.0006428801028608164,
      "grad_norm": 2.2097456455230713,
      "learning_rate": 4e-05,
      "loss": 7.2637,
      "step": 2
    },
    {
      "epoch": 0.0009643201542912247,
      "grad_norm": 1.8961241245269775,
      "learning_rate": 6e-05,
      "loss": 7.943,
      "step": 3
    },
    {
      "epoch": 0.0009643201542912247,
      "eval_loss": 2.014782428741455,
      "eval_runtime": 59.5967,
      "eval_samples_per_second": 21.981,
      "eval_steps_per_second": 10.991,
      "step": 3
    },
    {
      "epoch": 0.0012857602057216328,
      "grad_norm": 2.344888925552368,
      "learning_rate": 8e-05,
      "loss": 8.4286,
      "step": 4
    },
    {
      "epoch": 0.0016072002571520412,
      "grad_norm": 2.6145517826080322,
      "learning_rate": 0.0001,
      "loss": 9.0661,
      "step": 5
    },
    {
      "epoch": 0.0019286403085824494,
      "grad_norm": 1.9498443603515625,
      "learning_rate": 0.00012,
      "loss": 7.5982,
      "step": 6
    },
    {
      "epoch": 0.0019286403085824494,
      "eval_loss": 1.9811270236968994,
      "eval_runtime": 59.6772,
      "eval_samples_per_second": 21.951,
      "eval_steps_per_second": 10.976,
      "step": 6
    },
    {
      "epoch": 0.0022500803600128574,
      "grad_norm": 2.0184409618377686,
      "learning_rate": 0.00014,
      "loss": 7.381,
      "step": 7
    },
    {
      "epoch": 0.0025715204114432656,
      "grad_norm": 3.489924669265747,
      "learning_rate": 0.00016,
      "loss": 8.7444,
      "step": 8
    },
    {
      "epoch": 0.0028929604628736743,
      "grad_norm": 2.901017904281616,
      "learning_rate": 0.00018,
      "loss": 6.9573,
      "step": 9
    },
    {
      "epoch": 0.0028929604628736743,
      "eval_loss": 1.8749172687530518,
      "eval_runtime": 59.6544,
      "eval_samples_per_second": 21.96,
      "eval_steps_per_second": 10.98,
      "step": 9
    },
    {
      "epoch": 0.0032144005143040825,
      "grad_norm": 3.2114920616149902,
      "learning_rate": 0.0002,
      "loss": 7.9597,
      "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": 4065247635701760.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}