last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014534883720930232,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014534883720930232,
      "grad_norm": 0.5248333215713501,
      "learning_rate": 2e-05,
      "loss": 1.0081,
      "step": 1
    },
    {
      "epoch": 0.0014534883720930232,
      "eval_loss": 1.1038246154785156,
      "eval_runtime": 28.7768,
      "eval_samples_per_second": 5.039,
      "eval_steps_per_second": 5.039,
      "step": 1
    },
    {
      "epoch": 0.0029069767441860465,
      "grad_norm": 0.4121772050857544,
      "learning_rate": 4e-05,
      "loss": 0.8044,
      "step": 2
    },
    {
      "epoch": 0.00436046511627907,
      "grad_norm": 0.6754578948020935,
      "learning_rate": 6e-05,
      "loss": 1.402,
      "step": 3
    },
    {
      "epoch": 0.00436046511627907,
      "eval_loss": 1.0943968296051025,
      "eval_runtime": 39.0255,
      "eval_samples_per_second": 3.716,
      "eval_steps_per_second": 3.716,
      "step": 3
    },
    {
      "epoch": 0.005813953488372093,
      "grad_norm": 0.5914823412895203,
      "learning_rate": 8e-05,
      "loss": 0.8819,
      "step": 4
    },
    {
      "epoch": 0.007267441860465116,
      "grad_norm": 0.4640352129936218,
      "learning_rate": 0.0001,
      "loss": 0.9409,
      "step": 5
    },
    {
      "epoch": 0.00872093023255814,
      "grad_norm": 0.42091429233551025,
      "learning_rate": 0.00012,
      "loss": 0.9614,
      "step": 6
    },
    {
      "epoch": 0.00872093023255814,
      "eval_loss": 1.046032190322876,
      "eval_runtime": 48.9775,
      "eval_samples_per_second": 2.961,
      "eval_steps_per_second": 2.961,
      "step": 6
    },
    {
      "epoch": 0.010174418604651164,
      "grad_norm": 0.4285730719566345,
      "learning_rate": 0.00014,
      "loss": 1.0513,
      "step": 7
    },
    {
      "epoch": 0.011627906976744186,
      "grad_norm": 0.5643708109855652,
      "learning_rate": 0.00016,
      "loss": 3.323,
      "step": 8
    },
    {
      "epoch": 0.01308139534883721,
      "grad_norm": 0.37684959173202515,
      "learning_rate": 0.00018,
      "loss": 0.9996,
      "step": 9
    },
    {
      "epoch": 0.01308139534883721,
      "eval_loss": 0.9924975037574768,
      "eval_runtime": 48.9999,
      "eval_samples_per_second": 2.959,
      "eval_steps_per_second": 2.959,
      "step": 9
    },
    {
      "epoch": 0.014534883720930232,
      "grad_norm": 0.40745043754577637,
      "learning_rate": 0.0002,
      "loss": 0.9795,
      "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": 6534582019031040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}