last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005471207769115032,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005471207769115032,
      "grad_norm": 60.18626403808594,
      "learning_rate": 2e-05,
      "loss": 14.7588,
      "step": 1
    },
    {
      "epoch": 0.0005471207769115032,
      "eval_loss": 4.2187299728393555,
      "eval_runtime": 25.9522,
      "eval_samples_per_second": 29.67,
      "eval_steps_per_second": 14.835,
      "step": 1
    },
    {
      "epoch": 0.0010942415538230065,
      "grad_norm": 58.266685485839844,
      "learning_rate": 4e-05,
      "loss": 18.6801,
      "step": 2
    },
    {
      "epoch": 0.0016413623307345096,
      "grad_norm": 62.328731536865234,
      "learning_rate": 6e-05,
      "loss": 18.8466,
      "step": 3
    },
    {
      "epoch": 0.0016413623307345096,
      "eval_loss": 4.151100158691406,
      "eval_runtime": 24.5502,
      "eval_samples_per_second": 31.364,
      "eval_steps_per_second": 15.682,
      "step": 3
    },
    {
      "epoch": 0.002188483107646013,
      "grad_norm": 46.295589447021484,
      "learning_rate": 8e-05,
      "loss": 13.6519,
      "step": 4
    },
    {
      "epoch": 0.002735603884557516,
      "grad_norm": 70.07258605957031,
      "learning_rate": 0.0001,
      "loss": 15.6835,
      "step": 5
    },
    {
      "epoch": 0.003282724661469019,
      "grad_norm": 47.67641067504883,
      "learning_rate": 0.00012,
      "loss": 13.5726,
      "step": 6
    },
    {
      "epoch": 0.003282724661469019,
      "eval_loss": 3.4922778606414795,
      "eval_runtime": 24.709,
      "eval_samples_per_second": 31.163,
      "eval_steps_per_second": 15.581,
      "step": 6
    },
    {
      "epoch": 0.0038298454383805225,
      "grad_norm": 44.17360305786133,
      "learning_rate": 0.00014,
      "loss": 13.3029,
      "step": 7
    },
    {
      "epoch": 0.004376966215292026,
      "grad_norm": 46.18647384643555,
      "learning_rate": 0.00016,
      "loss": 12.597,
      "step": 8
    },
    {
      "epoch": 0.004924086992203529,
      "grad_norm": 53.092960357666016,
      "learning_rate": 0.00018,
      "loss": 13.4875,
      "step": 9
    },
    {
      "epoch": 0.004924086992203529,
      "eval_loss": 2.8043596744537354,
      "eval_runtime": 24.7368,
      "eval_samples_per_second": 31.128,
      "eval_steps_per_second": 15.564,
      "step": 9
    },
    {
      "epoch": 0.005471207769115032,
      "grad_norm": 60.76326370239258,
      "learning_rate": 0.0002,
      "loss": 10.2833,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1796487562395648.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}