last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018978333069745373,
  "eval_steps": 8,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006326111023248458,
      "eval_loss": 1.2243741750717163,
      "eval_runtime": 139.6436,
      "eval_samples_per_second": 4.769,
      "eval_steps_per_second": 2.385,
      "step": 1
    },
    {
      "epoch": 0.0018978333069745374,
      "grad_norm": 1.186109185218811,
      "learning_rate": 6e-05,
      "loss": 4.8593,
      "step": 3
    },
    {
      "epoch": 0.003795666613949075,
      "grad_norm": 1.3846362829208374,
      "learning_rate": 0.00012,
      "loss": 4.8476,
      "step": 6
    },
    {
      "epoch": 0.005060888818598766,
      "eval_loss": 1.1067157983779907,
      "eval_runtime": 141.5218,
      "eval_samples_per_second": 4.706,
      "eval_steps_per_second": 2.353,
      "step": 8
    },
    {
      "epoch": 0.005693499920923612,
      "grad_norm": 1.0471880435943604,
      "learning_rate": 0.00018,
      "loss": 4.558,
      "step": 9
    },
    {
      "epoch": 0.00759133322789815,
      "grad_norm": 1.3388583660125732,
      "learning_rate": 0.00019510565162951537,
      "loss": 4.236,
      "step": 12
    },
    {
      "epoch": 0.009489166534872686,
      "grad_norm": 1.3507390022277832,
      "learning_rate": 0.00017071067811865476,
      "loss": 3.9931,
      "step": 15
    },
    {
      "epoch": 0.010121777637197533,
      "eval_loss": 0.9410388469696045,
      "eval_runtime": 141.4592,
      "eval_samples_per_second": 4.708,
      "eval_steps_per_second": 2.354,
      "step": 16
    },
    {
      "epoch": 0.011386999841847224,
      "grad_norm": 1.206218957901001,
      "learning_rate": 0.00013090169943749476,
      "loss": 3.7541,
      "step": 18
    },
    {
      "epoch": 0.013284833148821762,
      "grad_norm": 1.0754749774932861,
      "learning_rate": 8.435655349597689e-05,
      "loss": 3.6349,
      "step": 21
    },
    {
      "epoch": 0.0151826664557963,
      "grad_norm": 1.0736587047576904,
      "learning_rate": 4.12214747707527e-05,
      "loss": 3.545,
      "step": 24
    },
    {
      "epoch": 0.0151826664557963,
      "eval_loss": 0.8697181940078735,
      "eval_runtime": 141.4788,
      "eval_samples_per_second": 4.707,
      "eval_steps_per_second": 2.354,
      "step": 24
    },
    {
      "epoch": 0.017080499762770835,
      "grad_norm": 0.9669196009635925,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 3.4764,
      "step": 27
    },
    {
      "epoch": 0.018978333069745373,
      "grad_norm": 0.9943149089813232,
      "learning_rate": 0.0,
      "loss": 3.4218,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 70,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.954479763095552e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}