last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00023673121537805974,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3673121537805975e-05,
      "grad_norm": 1.9601119756698608,
      "learning_rate": 2e-05,
      "loss": 1.5384,
      "step": 1
    },
    {
      "epoch": 2.3673121537805975e-05,
      "eval_loss": 1.7546149492263794,
      "eval_runtime": 498.5135,
      "eval_samples_per_second": 35.68,
      "eval_steps_per_second": 17.841,
      "step": 1
    },
    {
      "epoch": 4.734624307561195e-05,
      "grad_norm": 1.082641839981079,
      "learning_rate": 4e-05,
      "loss": 1.9956,
      "step": 2
    },
    {
      "epoch": 7.101936461341793e-05,
      "grad_norm": 1.5700252056121826,
      "learning_rate": 6e-05,
      "loss": 1.7681,
      "step": 3
    },
    {
      "epoch": 7.101936461341793e-05,
      "eval_loss": 1.745591640472412,
      "eval_runtime": 497.3239,
      "eval_samples_per_second": 35.765,
      "eval_steps_per_second": 17.884,
      "step": 3
    },
    {
      "epoch": 9.46924861512239e-05,
      "grad_norm": 1.921068787574768,
      "learning_rate": 8e-05,
      "loss": 1.6894,
      "step": 4
    },
    {
      "epoch": 0.00011836560768902987,
      "grad_norm": 1.6738073825836182,
      "learning_rate": 0.0001,
      "loss": 1.5135,
      "step": 5
    },
    {
      "epoch": 0.00014203872922683586,
      "grad_norm": 2.0782992839813232,
      "learning_rate": 0.00012,
      "loss": 1.9861,
      "step": 6
    },
    {
      "epoch": 0.00014203872922683586,
      "eval_loss": 1.6642322540283203,
      "eval_runtime": 497.4037,
      "eval_samples_per_second": 35.76,
      "eval_steps_per_second": 17.881,
      "step": 6
    },
    {
      "epoch": 0.0001657118507646418,
      "grad_norm": 1.8866004943847656,
      "learning_rate": 0.00014,
      "loss": 1.7562,
      "step": 7
    },
    {
      "epoch": 0.0001893849723024478,
      "grad_norm": 2.3774871826171875,
      "learning_rate": 0.00016,
      "loss": 1.8133,
      "step": 8
    },
    {
      "epoch": 0.00021305809384025378,
      "grad_norm": 2.0534048080444336,
      "learning_rate": 0.00018,
      "loss": 1.1913,
      "step": 9
    },
    {
      "epoch": 0.00021305809384025378,
      "eval_loss": 1.3677295446395874,
      "eval_runtime": 496.8183,
      "eval_samples_per_second": 35.802,
      "eval_steps_per_second": 17.902,
      "step": 9
    },
    {
      "epoch": 0.00023673121537805974,
      "grad_norm": 5.587805271148682,
      "learning_rate": 0.0002,
      "loss": 1.1873,
      "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": 1628733135912960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}