last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004064049418840933,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001625619767536373,
      "grad_norm": NaN,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0001625619767536373,
      "eval_loss": NaN,
      "eval_runtime": 302.4969,
      "eval_samples_per_second": 8.565,
      "eval_steps_per_second": 2.142,
      "step": 1
    },
    {
      "epoch": 0.0003251239535072746,
      "grad_norm": NaN,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.000487685930260912,
      "grad_norm": NaN,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0006502479070145493,
      "grad_norm": NaN,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0006502479070145493,
      "eval_loss": NaN,
      "eval_runtime": 300.0714,
      "eval_samples_per_second": 8.635,
      "eval_steps_per_second": 2.159,
      "step": 4
    },
    {
      "epoch": 0.0008128098837681866,
      "grad_norm": NaN,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.000975371860521824,
      "grad_norm": NaN,
      "learning_rate": 6e-06,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0011379338372754612,
      "grad_norm": NaN,
      "learning_rate": 7e-06,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0013004958140290985,
      "grad_norm": NaN,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0013004958140290985,
      "eval_loss": NaN,
      "eval_runtime": 301.324,
      "eval_samples_per_second": 8.599,
      "eval_steps_per_second": 2.151,
      "step": 8
    },
    {
      "epoch": 0.001463057790782736,
      "grad_norm": NaN,
      "learning_rate": 9e-06,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0016256197675363732,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0017881817442900105,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.001950743721043648,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-06,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.001950743721043648,
      "eval_loss": NaN,
      "eval_runtime": 300.4107,
      "eval_samples_per_second": 8.625,
      "eval_steps_per_second": 2.157,
      "step": 12
    },
    {
      "epoch": 0.002113305697797285,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0022758676745509225,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.00243842965130456,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.002600991628058197,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.002600991628058197,
      "eval_loss": NaN,
      "eval_runtime": 300.7597,
      "eval_samples_per_second": 8.615,
      "eval_steps_per_second": 2.155,
      "step": 16
    },
    {
      "epoch": 0.0027635536048118345,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.002926115581565472,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.003088677558319109,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0032512395350727465,
      "grad_norm": NaN,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0032512395350727465,
      "eval_loss": NaN,
      "eval_runtime": 300.5225,
      "eval_samples_per_second": 8.622,
      "eval_steps_per_second": 2.156,
      "step": 20
    },
    {
      "epoch": 0.003413801511826384,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.003576363488580021,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.0037389254653336585,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-07,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.003901487442087296,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.003901487442087296,
      "eval_loss": NaN,
      "eval_runtime": 300.7061,
      "eval_samples_per_second": 8.616,
      "eval_steps_per_second": 2.155,
      "step": 24
    },
    {
      "epoch": 0.004064049418840933,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9273591044505600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}