last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00011256539345826216,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1256539345826217e-05,
      "grad_norm": 4.521182537078857,
      "learning_rate": 2e-05,
      "loss": 17.1448,
      "step": 1
    },
    {
      "epoch": 1.1256539345826217e-05,
      "eval_loss": 4.492743015289307,
      "eval_runtime": 2081.7009,
      "eval_samples_per_second": 17.969,
      "eval_steps_per_second": 8.984,
      "step": 1
    },
    {
      "epoch": 2.2513078691652433e-05,
      "grad_norm": 10.29865550994873,
      "learning_rate": 4e-05,
      "loss": 19.7466,
      "step": 2
    },
    {
      "epoch": 3.3769618037478645e-05,
      "grad_norm": 4.724206447601318,
      "learning_rate": 6e-05,
      "loss": 18.6372,
      "step": 3
    },
    {
      "epoch": 3.3769618037478645e-05,
      "eval_loss": 4.4801788330078125,
      "eval_runtime": 2074.5771,
      "eval_samples_per_second": 18.031,
      "eval_steps_per_second": 9.015,
      "step": 3
    },
    {
      "epoch": 4.502615738330487e-05,
      "grad_norm": 8.207385063171387,
      "learning_rate": 8e-05,
      "loss": 17.332,
      "step": 4
    },
    {
      "epoch": 5.628269672913108e-05,
      "grad_norm": 6.5473456382751465,
      "learning_rate": 0.0001,
      "loss": 18.4788,
      "step": 5
    },
    {
      "epoch": 6.753923607495729e-05,
      "grad_norm": 7.216912269592285,
      "learning_rate": 0.00012,
      "loss": 17.2112,
      "step": 6
    },
    {
      "epoch": 6.753923607495729e-05,
      "eval_loss": 4.203959941864014,
      "eval_runtime": 2080.6001,
      "eval_samples_per_second": 17.978,
      "eval_steps_per_second": 8.989,
      "step": 6
    },
    {
      "epoch": 7.879577542078351e-05,
      "grad_norm": 6.758055686950684,
      "learning_rate": 0.00014,
      "loss": 20.4907,
      "step": 7
    },
    {
      "epoch": 9.005231476660973e-05,
      "grad_norm": 8.21561050415039,
      "learning_rate": 0.00016,
      "loss": 14.2335,
      "step": 8
    },
    {
      "epoch": 0.00010130885411243594,
      "grad_norm": 10.57795524597168,
      "learning_rate": 0.00018,
      "loss": 17.5631,
      "step": 9
    },
    {
      "epoch": 0.00010130885411243594,
      "eval_loss": 3.2738890647888184,
      "eval_runtime": 2094.3921,
      "eval_samples_per_second": 17.86,
      "eval_steps_per_second": 8.93,
      "step": 9
    },
    {
      "epoch": 0.00011256539345826216,
      "grad_norm": 11.34732437133789,
      "learning_rate": 0.0002,
      "loss": 13.204,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}