last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1949317738791423,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003898635477582846,
      "eval_loss": NaN,
      "eval_runtime": 12.9048,
      "eval_samples_per_second": 8.369,
      "eval_steps_per_second": 4.184,
      "step": 1
    },
    {
      "epoch": 0.01949317738791423,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.03898635477582846,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.03898635477582846,
      "eval_loss": NaN,
      "eval_runtime": 12.4762,
      "eval_samples_per_second": 8.656,
      "eval_steps_per_second": 4.328,
      "step": 10
    },
    {
      "epoch": 0.05847953216374269,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.07797270955165692,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.07797270955165692,
      "eval_loss": NaN,
      "eval_runtime": 12.7644,
      "eval_samples_per_second": 8.461,
      "eval_steps_per_second": 4.231,
      "step": 20
    },
    {
      "epoch": 0.09746588693957114,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.11695906432748537,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.11695906432748537,
      "eval_loss": NaN,
      "eval_runtime": 12.9092,
      "eval_samples_per_second": 8.366,
      "eval_steps_per_second": 4.183,
      "step": 30
    },
    {
      "epoch": 0.1364522417153996,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0173,
      "step": 35
    },
    {
      "epoch": 0.15594541910331383,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.15594541910331383,
      "eval_loss": NaN,
      "eval_runtime": 12.9787,
      "eval_samples_per_second": 8.321,
      "eval_steps_per_second": 4.161,
      "step": 40
    },
    {
      "epoch": 0.17543859649122806,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0068,
      "step": 45
    },
    {
      "epoch": 0.1949317738791423,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0161,
      "step": 50
    },
    {
      "epoch": 0.1949317738791423,
      "eval_loss": NaN,
      "eval_runtime": 13.0746,
      "eval_samples_per_second": 8.26,
      "eval_steps_per_second": 4.13,
      "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": 1.2939072642220032e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}