last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.027387880862718247,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018258587241812164,
      "grad_norm": 1.3482320308685303,
      "learning_rate": 1e-05,
      "loss": 1.9459,
      "step": 1
    },
    {
      "epoch": 0.0018258587241812164,
      "eval_loss": 1.9964624643325806,
      "eval_runtime": 170.8937,
      "eval_samples_per_second": 5.401,
      "eval_steps_per_second": 2.703,
      "step": 1
    },
    {
      "epoch": 0.0036517174483624328,
      "grad_norm": 1.172255039215088,
      "learning_rate": 2e-05,
      "loss": 1.8797,
      "step": 2
    },
    {
      "epoch": 0.005477576172543649,
      "grad_norm": 1.1005375385284424,
      "learning_rate": 3e-05,
      "loss": 1.8084,
      "step": 3
    },
    {
      "epoch": 0.0073034348967248655,
      "grad_norm": 1.1677844524383545,
      "learning_rate": 4e-05,
      "loss": 1.7838,
      "step": 4
    },
    {
      "epoch": 0.009129293620906082,
      "grad_norm": 1.6915104389190674,
      "learning_rate": 5e-05,
      "loss": 1.9926,
      "step": 5
    },
    {
      "epoch": 0.009129293620906082,
      "eval_loss": 1.9939281940460205,
      "eval_runtime": 13.5369,
      "eval_samples_per_second": 68.184,
      "eval_steps_per_second": 34.129,
      "step": 5
    },
    {
      "epoch": 0.010955152345087298,
      "grad_norm": 1.3168697357177734,
      "learning_rate": 6e-05,
      "loss": 2.0256,
      "step": 6
    },
    {
      "epoch": 0.012781011069268515,
      "grad_norm": 1.17354416847229,
      "learning_rate": 7e-05,
      "loss": 1.9324,
      "step": 7
    },
    {
      "epoch": 0.014606869793449731,
      "grad_norm": 1.2946925163269043,
      "learning_rate": 8e-05,
      "loss": 1.8946,
      "step": 8
    },
    {
      "epoch": 0.01643272851763095,
      "grad_norm": 1.331744909286499,
      "learning_rate": 9e-05,
      "loss": 2.0392,
      "step": 9
    },
    {
      "epoch": 0.018258587241812164,
      "grad_norm": 1.261064887046814,
      "learning_rate": 0.0001,
      "loss": 1.9267,
      "step": 10
    },
    {
      "epoch": 0.018258587241812164,
      "eval_loss": 1.9822499752044678,
      "eval_runtime": 13.5702,
      "eval_samples_per_second": 68.017,
      "eval_steps_per_second": 34.045,
      "step": 10
    },
    {
      "epoch": 0.020084445965993382,
      "grad_norm": 1.2413709163665771,
      "learning_rate": 9.755282581475769e-05,
      "loss": 1.9181,
      "step": 11
    },
    {
      "epoch": 0.021910304690174597,
      "grad_norm": 1.4612451791763306,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.0087,
      "step": 12
    },
    {
      "epoch": 0.023736163414355815,
      "grad_norm": 1.5164417028427124,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.7626,
      "step": 13
    },
    {
      "epoch": 0.02556202213853703,
      "grad_norm": 2.2981152534484863,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.9643,
      "step": 14
    },
    {
      "epoch": 0.027387880862718247,
      "grad_norm": 1.6825339794158936,
      "learning_rate": 5e-05,
      "loss": 1.8795,
      "step": 15
    },
    {
      "epoch": 0.027387880862718247,
      "eval_loss": 1.9565916061401367,
      "eval_runtime": 13.5223,
      "eval_samples_per_second": 68.258,
      "eval_steps_per_second": 34.166,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4331570410291200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}