last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.020092425155716295,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004018485031143259,
      "eval_loss": NaN,
      "eval_runtime": 12.1629,
      "eval_samples_per_second": 86.164,
      "eval_steps_per_second": 43.082,
      "step": 1
    },
    {
      "epoch": 0.0020092425155716293,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.004018485031143259,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.004018485031143259,
      "eval_loss": NaN,
      "eval_runtime": 11.3335,
      "eval_samples_per_second": 92.469,
      "eval_steps_per_second": 46.235,
      "step": 10
    },
    {
      "epoch": 0.006027727546714889,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.008036970062286517,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.008036970062286517,
      "eval_loss": NaN,
      "eval_runtime": 11.3286,
      "eval_samples_per_second": 92.509,
      "eval_steps_per_second": 46.255,
      "step": 20
    },
    {
      "epoch": 0.010046212577858147,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.012055455093429777,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.012055455093429777,
      "eval_loss": NaN,
      "eval_runtime": 11.2266,
      "eval_samples_per_second": 93.35,
      "eval_steps_per_second": 46.675,
      "step": 30
    },
    {
      "epoch": 0.014064697609001406,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.016073940124573034,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.016073940124573034,
      "eval_loss": NaN,
      "eval_runtime": 11.3364,
      "eval_samples_per_second": 92.445,
      "eval_steps_per_second": 46.223,
      "step": 40
    },
    {
      "epoch": 0.018083182640144666,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.020092425155716295,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.020092425155716295,
      "eval_loss": NaN,
      "eval_runtime": 11.2152,
      "eval_samples_per_second": 93.444,
      "eval_steps_per_second": 46.722,
      "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": 410237708009472.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}