last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001505253334136135,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.0105066682722702e-05,
      "eval_loss": NaN,
      "eval_runtime": 1573.8893,
      "eval_samples_per_second": 8.887,
      "eval_steps_per_second": 4.444,
      "step": 1
    },
    {
      "epoch": 0.00015052533341361352,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00030105066682722703,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.00030105066682722703,
      "eval_loss": NaN,
      "eval_runtime": 1571.1396,
      "eval_samples_per_second": 8.902,
      "eval_steps_per_second": 4.452,
      "step": 10
    },
    {
      "epoch": 0.0004515760002408405,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0006021013336544541,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0006021013336544541,
      "eval_loss": NaN,
      "eval_runtime": 1571.5208,
      "eval_samples_per_second": 8.9,
      "eval_steps_per_second": 4.45,
      "step": 20
    },
    {
      "epoch": 0.0007526266670680675,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.000903152000481681,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.000903152000481681,
      "eval_loss": NaN,
      "eval_runtime": 1571.7037,
      "eval_samples_per_second": 8.899,
      "eval_steps_per_second": 4.45,
      "step": 30
    },
    {
      "epoch": 0.0010536773338952945,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.0012042026673089081,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.0012042026673089081,
      "eval_loss": NaN,
      "eval_runtime": 1571.1375,
      "eval_samples_per_second": 8.902,
      "eval_steps_per_second": 4.452,
      "step": 40
    },
    {
      "epoch": 0.0013547280007225216,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.001505253334136135,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.001505253334136135,
      "eval_loss": NaN,
      "eval_runtime": 1569.7303,
      "eval_samples_per_second": 8.91,
      "eval_steps_per_second": 4.456,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7077833251684352e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}