last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0026139946739858516,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00026139946739858516,
      "grad_norm": 0.8078809380531311,
      "learning_rate": 1e-05,
      "loss": 0.3212,
      "step": 1
    },
    {
      "epoch": 0.00026139946739858516,
      "eval_loss": 0.4257713258266449,
      "eval_runtime": 1041.9475,
      "eval_samples_per_second": 6.184,
      "eval_steps_per_second": 3.092,
      "step": 1
    },
    {
      "epoch": 0.0005227989347971703,
      "grad_norm": 0.7678937315940857,
      "learning_rate": 2e-05,
      "loss": 0.2916,
      "step": 2
    },
    {
      "epoch": 0.0007841984021957555,
      "grad_norm": 0.8585813641548157,
      "learning_rate": 3e-05,
      "loss": 0.3591,
      "step": 3
    },
    {
      "epoch": 0.0007841984021957555,
      "eval_loss": NaN,
      "eval_runtime": 1019.3375,
      "eval_samples_per_second": 6.321,
      "eval_steps_per_second": 3.161,
      "step": 3
    },
    {
      "epoch": 0.0010455978695943407,
      "grad_norm": 0.9703998565673828,
      "learning_rate": 4e-05,
      "loss": 0.3465,
      "step": 4
    },
    {
      "epoch": 0.0013069973369929258,
      "grad_norm": 0.7693590521812439,
      "learning_rate": 5e-05,
      "loss": 0.2594,
      "step": 5
    },
    {
      "epoch": 0.001568396804391511,
      "grad_norm": 0.8005192279815674,
      "learning_rate": 6e-05,
      "loss": 0.2363,
      "step": 6
    },
    {
      "epoch": 0.001568396804391511,
      "eval_loss": NaN,
      "eval_runtime": 1020.151,
      "eval_samples_per_second": 6.316,
      "eval_steps_per_second": 3.158,
      "step": 6
    },
    {
      "epoch": 0.0018297962717900961,
      "grad_norm": 0.7309572696685791,
      "learning_rate": 7e-05,
      "loss": 0.1376,
      "step": 7
    },
    {
      "epoch": 0.0020911957391886813,
      "grad_norm": 1.0636768341064453,
      "learning_rate": 8e-05,
      "loss": 0.1193,
      "step": 8
    },
    {
      "epoch": 0.0023525952065872667,
      "grad_norm": 0.42810970544815063,
      "learning_rate": 9e-05,
      "loss": 0.0449,
      "step": 9
    },
    {
      "epoch": 0.0023525952065872667,
      "eval_loss": NaN,
      "eval_runtime": 1020.1784,
      "eval_samples_per_second": 6.316,
      "eval_steps_per_second": 3.158,
      "step": 9
    },
    {
      "epoch": 0.0026139946739858516,
      "grad_norm": 0.7571702599525452,
      "learning_rate": 0.0001,
      "loss": 0.0312,
      "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": 5.972192632661606e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}