last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0006319398182646406,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.2129321217642704e-05,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 4.2129321217642704e-05,
      "eval_loss": NaN,
      "eval_runtime": 1966.7409,
      "eval_samples_per_second": 20.327,
      "eval_steps_per_second": 10.164,
      "step": 1
    },
    {
      "epoch": 8.425864243528541e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00012638796365292812,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00016851728487057082,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.00021064660608821354,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00021064660608821354,
      "eval_loss": NaN,
      "eval_runtime": 1699.947,
      "eval_samples_per_second": 23.517,
      "eval_steps_per_second": 11.759,
      "step": 5
    },
    {
      "epoch": 0.00025277592730585624,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00029490524852349896,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.00033703456974114163,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.00037916389095878436,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0004212932121764271,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0004212932121764271,
      "eval_loss": NaN,
      "eval_runtime": 2321.372,
      "eval_samples_per_second": 17.222,
      "eval_steps_per_second": 8.611,
      "step": 10
    },
    {
      "epoch": 0.00046342253339406975,
      "grad_norm": NaN,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0005055518546117125,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 6461.936,
      "step": 12
    },
    {
      "epoch": 0.0005476811758293551,
      "grad_norm": NaN,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0005898104970469979,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0006319398182646406,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0006319398182646406,
      "eval_loss": NaN,
      "eval_runtime": 3326.0425,
      "eval_samples_per_second": 12.02,
      "eval_steps_per_second": 6.01,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.156552165687296e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}