last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007661250162801566,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.5322500325603132e-05,
      "eval_loss": NaN,
      "eval_runtime": 1265.0534,
      "eval_samples_per_second": 21.722,
      "eval_steps_per_second": 10.861,
      "step": 1
    },
    {
      "epoch": 7.661250162801566e-05,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 29.8212,
      "step": 5
    },
    {
      "epoch": 0.00015322500325603132,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.00015322500325603132,
      "eval_loss": NaN,
      "eval_runtime": 1265.6774,
      "eval_samples_per_second": 21.712,
      "eval_steps_per_second": 10.856,
      "step": 10
    },
    {
      "epoch": 0.00022983750488404697,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.00030645000651206264,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.00030645000651206264,
      "eval_loss": NaN,
      "eval_runtime": 1269.5948,
      "eval_samples_per_second": 21.645,
      "eval_steps_per_second": 10.822,
      "step": 20
    },
    {
      "epoch": 0.0003830625081400783,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 8.4151,
      "step": 25
    },
    {
      "epoch": 0.00045967500976809393,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.00045967500976809393,
      "eval_loss": NaN,
      "eval_runtime": 1268.5609,
      "eval_samples_per_second": 21.662,
      "eval_steps_per_second": 10.831,
      "step": 30
    },
    {
      "epoch": 0.0005362875113961096,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.0006129000130241253,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.0006129000130241253,
      "eval_loss": NaN,
      "eval_runtime": 1265.8644,
      "eval_samples_per_second": 21.708,
      "eval_steps_per_second": 10.854,
      "step": 40
    },
    {
      "epoch": 0.000689512514652141,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.0007661250162801566,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.0007661250162801566,
      "eval_loss": NaN,
      "eval_runtime": 1266.8065,
      "eval_samples_per_second": 21.692,
      "eval_steps_per_second": 10.846,
      "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": 5281565890314240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}