last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0010654733365297534,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010654733365297533,
      "grad_norm": 1.6343660354614258,
      "learning_rate": 2e-05,
      "loss": 2.9671,
      "step": 1
    },
    {
      "epoch": 0.00010654733365297533,
      "eval_loss": 3.0384771823883057,
      "eval_runtime": 125.6181,
      "eval_samples_per_second": 31.46,
      "eval_steps_per_second": 15.73,
      "step": 1
    },
    {
      "epoch": 0.00021309466730595067,
      "grad_norm": 1.2648301124572754,
      "learning_rate": 4e-05,
      "loss": 2.8694,
      "step": 2
    },
    {
      "epoch": 0.000319642000958926,
      "grad_norm": 0.9710222482681274,
      "learning_rate": 6e-05,
      "loss": 2.8004,
      "step": 3
    },
    {
      "epoch": 0.000319642000958926,
      "eval_loss": 3.0328028202056885,
      "eval_runtime": 124.0247,
      "eval_samples_per_second": 31.865,
      "eval_steps_per_second": 15.932,
      "step": 3
    },
    {
      "epoch": 0.00042618933461190133,
      "grad_norm": 1.5170894861221313,
      "learning_rate": 8e-05,
      "loss": 2.9305,
      "step": 4
    },
    {
      "epoch": 0.0005327366682648767,
      "grad_norm": 1.787522792816162,
      "learning_rate": 0.0001,
      "loss": 3.1154,
      "step": 5
    },
    {
      "epoch": 0.000639284001917852,
      "grad_norm": 1.872839331626892,
      "learning_rate": 0.00012,
      "loss": 2.7988,
      "step": 6
    },
    {
      "epoch": 0.000639284001917852,
      "eval_loss": 2.9444522857666016,
      "eval_runtime": 124.464,
      "eval_samples_per_second": 31.752,
      "eval_steps_per_second": 15.876,
      "step": 6
    },
    {
      "epoch": 0.0007458313355708274,
      "grad_norm": 1.8387773036956787,
      "learning_rate": 0.00014,
      "loss": 3.155,
      "step": 7
    },
    {
      "epoch": 0.0008523786692238027,
      "grad_norm": 1.3405261039733887,
      "learning_rate": 0.00016,
      "loss": 2.5581,
      "step": 8
    },
    {
      "epoch": 0.000958926002876778,
      "grad_norm": 2.029334306716919,
      "learning_rate": 0.00018,
      "loss": 2.6493,
      "step": 9
    },
    {
      "epoch": 0.000958926002876778,
      "eval_loss": 2.722989082336426,
      "eval_runtime": 127.4043,
      "eval_samples_per_second": 31.019,
      "eval_steps_per_second": 15.51,
      "step": 9
    },
    {
      "epoch": 0.0010654733365297534,
      "grad_norm": 1.4754557609558105,
      "learning_rate": 0.0002,
      "loss": 2.6873,
      "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
}