last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00210737052842316,
  "eval_steps": 13,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.21474105684632e-05,
      "eval_loss": 0.6331890225410461,
      "eval_runtime": 784.169,
      "eval_samples_per_second": 12.74,
      "eval_steps_per_second": 6.37,
      "step": 1
    },
    {
      "epoch": 0.000210737052842316,
      "grad_norm": 4.441408634185791,
      "learning_rate": 5e-05,
      "loss": 2.8264,
      "step": 5
    },
    {
      "epoch": 0.000421474105684632,
      "grad_norm": 3.703730583190918,
      "learning_rate": 0.0001,
      "loss": 2.2459,
      "step": 10
    },
    {
      "epoch": 0.0005479163373900216,
      "eval_loss": 0.5664745569229126,
      "eval_runtime": 787.5421,
      "eval_samples_per_second": 12.685,
      "eval_steps_per_second": 6.343,
      "step": 13
    },
    {
      "epoch": 0.000632211158526948,
      "grad_norm": 3.4340572357177734,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.2234,
      "step": 15
    },
    {
      "epoch": 0.000842948211369264,
      "grad_norm": 2.968557119369507,
      "learning_rate": 8.535533905932738e-05,
      "loss": 2.156,
      "step": 20
    },
    {
      "epoch": 0.00105368526421158,
      "grad_norm": 3.138040065765381,
      "learning_rate": 6.91341716182545e-05,
      "loss": 1.9719,
      "step": 25
    },
    {
      "epoch": 0.0010958326747800432,
      "eval_loss": 0.5203956961631775,
      "eval_runtime": 789.5111,
      "eval_samples_per_second": 12.653,
      "eval_steps_per_second": 6.327,
      "step": 26
    },
    {
      "epoch": 0.001264422317053896,
      "grad_norm": 3.788905382156372,
      "learning_rate": 5e-05,
      "loss": 2.0411,
      "step": 30
    },
    {
      "epoch": 0.001475159369896212,
      "grad_norm": 3.120276689529419,
      "learning_rate": 3.086582838174551e-05,
      "loss": 2.0862,
      "step": 35
    },
    {
      "epoch": 0.0016437490121700648,
      "eval_loss": 0.5087147951126099,
      "eval_runtime": 789.1636,
      "eval_samples_per_second": 12.659,
      "eval_steps_per_second": 6.329,
      "step": 39
    },
    {
      "epoch": 0.001685896422738528,
      "grad_norm": 2.999903917312622,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 1.807,
      "step": 40
    },
    {
      "epoch": 0.001896633475580844,
      "grad_norm": 3.139343500137329,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 2.2568,
      "step": 45
    },
    {
      "epoch": 0.00210737052842316,
      "grad_norm": 3.2525317668914795,
      "learning_rate": 0.0,
      "loss": 1.948,
      "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": 1.708853041299456e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}