last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0006168841183183739,
  "eval_steps": 2,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.0844205915918693e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 3.0844205915918693e-05,
      "eval_loss": NaN,
      "eval_runtime": 468.8432,
      "eval_samples_per_second": 29.116,
      "eval_steps_per_second": 14.559,
      "step": 1
    },
    {
      "epoch": 6.168841183183739e-05,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 6.168841183183739e-05,
      "eval_loss": NaN,
      "eval_runtime": 467.4001,
      "eval_samples_per_second": 29.206,
      "eval_steps_per_second": 14.604,
      "step": 2
    },
    {
      "epoch": 9.253261774775608e-05,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00012337682366367477,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.00012337682366367477,
      "eval_loss": NaN,
      "eval_runtime": 467.7941,
      "eval_samples_per_second": 29.182,
      "eval_steps_per_second": 14.592,
      "step": 4
    },
    {
      "epoch": 0.00015422102957959348,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00018506523549551216,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00018506523549551216,
      "eval_loss": NaN,
      "eval_runtime": 467.5617,
      "eval_samples_per_second": 29.196,
      "eval_steps_per_second": 14.599,
      "step": 6
    },
    {
      "epoch": 0.00021590944141143087,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.00024675364732734955,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.00024675364732734955,
      "eval_loss": NaN,
      "eval_runtime": 467.3563,
      "eval_samples_per_second": 29.209,
      "eval_steps_per_second": 14.606,
      "step": 8
    },
    {
      "epoch": 0.0002775978532432682,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.00030844205915918696,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.00030844205915918696,
      "eval_loss": NaN,
      "eval_runtime": 467.3596,
      "eval_samples_per_second": 29.209,
      "eval_steps_per_second": 14.605,
      "step": 10
    },
    {
      "epoch": 0.00033928626507510564,
      "grad_norm": NaN,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0003701304709910243,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0003701304709910243,
      "eval_loss": NaN,
      "eval_runtime": 467.4243,
      "eval_samples_per_second": 29.205,
      "eval_steps_per_second": 14.603,
      "step": 12
    },
    {
      "epoch": 0.00040097467690694305,
      "grad_norm": NaN,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.00043181888282286173,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.00043181888282286173,
      "eval_loss": NaN,
      "eval_runtime": 467.5965,
      "eval_samples_per_second": 29.194,
      "eval_steps_per_second": 14.598,
      "step": 14
    },
    {
      "epoch": 0.0004626630887387804,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0004935072946546991,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.0004935072946546991,
      "eval_loss": NaN,
      "eval_runtime": 467.5922,
      "eval_samples_per_second": 29.194,
      "eval_steps_per_second": 14.598,
      "step": 16
    },
    {
      "epoch": 0.0005243515005706178,
      "grad_norm": NaN,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.0005551957064865365,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0005551957064865365,
      "eval_loss": NaN,
      "eval_runtime": 467.1377,
      "eval_samples_per_second": 29.223,
      "eval_steps_per_second": 14.612,
      "step": 18
    },
    {
      "epoch": 0.0005860399124024552,
      "grad_norm": NaN,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0006168841183183739,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0006168841183183739,
      "eval_loss": NaN,
      "eval_runtime": 467.1363,
      "eval_samples_per_second": 29.223,
      "eval_steps_per_second": 14.612,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2046335826001920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}