last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003372169485238328,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.372169485238328e-05,
      "grad_norm": 0.3184521794319153,
      "learning_rate": 2e-05,
      "loss": 2.6964,
      "step": 1
    },
    {
      "epoch": 3.372169485238328e-05,
      "eval_loss": 0.6847627758979797,
      "eval_runtime": 860.4421,
      "eval_samples_per_second": 14.512,
      "eval_steps_per_second": 7.257,
      "step": 1
    },
    {
      "epoch": 6.744338970476656e-05,
      "grad_norm": 0.2911832332611084,
      "learning_rate": 4e-05,
      "loss": 2.5339,
      "step": 2
    },
    {
      "epoch": 0.00010116508455714984,
      "grad_norm": 0.35111960768699646,
      "learning_rate": 6e-05,
      "loss": 2.7398,
      "step": 3
    },
    {
      "epoch": 0.00010116508455714984,
      "eval_loss": 0.6845397353172302,
      "eval_runtime": 860.6875,
      "eval_samples_per_second": 14.508,
      "eval_steps_per_second": 7.255,
      "step": 3
    },
    {
      "epoch": 0.0001348867794095331,
      "grad_norm": 0.3350541591644287,
      "learning_rate": 8e-05,
      "loss": 2.5892,
      "step": 4
    },
    {
      "epoch": 0.0001686084742619164,
      "grad_norm": 0.31926605105400085,
      "learning_rate": 0.0001,
      "loss": 2.8168,
      "step": 5
    },
    {
      "epoch": 0.00020233016911429968,
      "grad_norm": 0.345554918050766,
      "learning_rate": 0.00012,
      "loss": 2.592,
      "step": 6
    },
    {
      "epoch": 0.00020233016911429968,
      "eval_loss": 0.6812539100646973,
      "eval_runtime": 861.2331,
      "eval_samples_per_second": 14.499,
      "eval_steps_per_second": 7.25,
      "step": 6
    },
    {
      "epoch": 0.00023605186396668295,
      "grad_norm": 0.41131940484046936,
      "learning_rate": 0.00014,
      "loss": 2.7157,
      "step": 7
    },
    {
      "epoch": 0.0002697735588190662,
      "grad_norm": 0.45540064573287964,
      "learning_rate": 0.00016,
      "loss": 2.9485,
      "step": 8
    },
    {
      "epoch": 0.00030349525367144955,
      "grad_norm": 0.5577786564826965,
      "learning_rate": 0.00018,
      "loss": 3.2718,
      "step": 9
    },
    {
      "epoch": 0.00030349525367144955,
      "eval_loss": 0.6599324941635132,
      "eval_runtime": 861.0508,
      "eval_samples_per_second": 14.502,
      "eval_steps_per_second": 7.252,
      "step": 9
    },
    {
      "epoch": 0.0003372169485238328,
      "grad_norm": 0.6703528761863708,
      "learning_rate": 0.0002,
      "loss": 3.3476,
      "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": 3257498484080640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}