trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.959016393442623,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32786885245901637,
      "grad_norm": 17.434409581805603,
      "learning_rate": 5e-06,
      "loss": 1.1821,
      "step": 10
    },
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 7.59380268216057,
      "learning_rate": 5e-06,
      "loss": 1.0722,
      "step": 20
    },
    {
      "epoch": 0.9836065573770492,
      "grad_norm": 1.5863650387592914,
      "learning_rate": 5e-06,
      "loss": 1.0317,
      "step": 30
    },
    {
      "epoch": 0.9836065573770492,
      "eval_loss": 1.0128474235534668,
      "eval_runtime": 32.8639,
      "eval_samples_per_second": 24.982,
      "eval_steps_per_second": 0.396,
      "step": 30
    },
    {
      "epoch": 1.3155737704918034,
      "grad_norm": 2.3139241529035615,
      "learning_rate": 5e-06,
      "loss": 1.0379,
      "step": 40
    },
    {
      "epoch": 1.6434426229508197,
      "grad_norm": 1.0834656481897813,
      "learning_rate": 5e-06,
      "loss": 0.952,
      "step": 50
    },
    {
      "epoch": 1.971311475409836,
      "grad_norm": 3.294871252924834,
      "learning_rate": 5e-06,
      "loss": 0.942,
      "step": 60
    },
    {
      "epoch": 1.971311475409836,
      "eval_loss": 0.966790497303009,
      "eval_runtime": 32.8378,
      "eval_samples_per_second": 25.002,
      "eval_steps_per_second": 0.396,
      "step": 60
    },
    {
      "epoch": 2.30327868852459,
      "grad_norm": 2.469625552209217,
      "learning_rate": 5e-06,
      "loss": 0.9525,
      "step": 70
    },
    {
      "epoch": 2.6311475409836067,
      "grad_norm": 1.8447831909454038,
      "learning_rate": 5e-06,
      "loss": 0.8794,
      "step": 80
    },
    {
      "epoch": 2.959016393442623,
      "grad_norm": 1.9087591460647273,
      "learning_rate": 5e-06,
      "loss": 0.8784,
      "step": 90
    },
    {
      "epoch": 2.959016393442623,
      "eval_loss": 0.9442600607872009,
      "eval_runtime": 31.7685,
      "eval_samples_per_second": 25.843,
      "eval_steps_per_second": 0.409,
      "step": 90
    },
    {
      "epoch": 2.959016393442623,
      "step": 90,
      "total_flos": 150543972433920.0,
      "train_loss": 0.9920466634962294,
      "train_runtime": 5576.9487,
      "train_samples_per_second": 8.385,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 10,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 150543972433920.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}