last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006021375884389583,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006021375884389584,
      "grad_norm": 0.006295036058872938,
      "learning_rate": 2e-05,
      "loss": 12.4525,
      "step": 1
    },
    {
      "epoch": 0.0006021375884389584,
      "eval_loss": 12.454846382141113,
      "eval_runtime": 6.8267,
      "eval_samples_per_second": 51.269,
      "eval_steps_per_second": 51.269,
      "step": 1
    },
    {
      "epoch": 0.0012042751768779167,
      "grad_norm": 0.006132213864475489,
      "learning_rate": 4e-05,
      "loss": 12.4598,
      "step": 2
    },
    {
      "epoch": 0.0018064127653168748,
      "grad_norm": 0.00741590978577733,
      "learning_rate": 6e-05,
      "loss": 12.4586,
      "step": 3
    },
    {
      "epoch": 0.0018064127653168748,
      "eval_loss": 12.454843521118164,
      "eval_runtime": 6.814,
      "eval_samples_per_second": 51.365,
      "eval_steps_per_second": 51.365,
      "step": 3
    },
    {
      "epoch": 0.0024085503537558334,
      "grad_norm": 0.006793172564357519,
      "learning_rate": 8e-05,
      "loss": 12.452,
      "step": 4
    },
    {
      "epoch": 0.0030106879421947915,
      "grad_norm": 0.005385817494243383,
      "learning_rate": 0.0001,
      "loss": 12.4524,
      "step": 5
    },
    {
      "epoch": 0.0036128255306337497,
      "grad_norm": 0.008517769165337086,
      "learning_rate": 0.00012,
      "loss": 12.4596,
      "step": 6
    },
    {
      "epoch": 0.0036128255306337497,
      "eval_loss": 12.454833030700684,
      "eval_runtime": 6.8407,
      "eval_samples_per_second": 51.165,
      "eval_steps_per_second": 51.165,
      "step": 6
    },
    {
      "epoch": 0.004214963119072708,
      "grad_norm": 0.0068705081939697266,
      "learning_rate": 0.00014,
      "loss": 12.4564,
      "step": 7
    },
    {
      "epoch": 0.004817100707511667,
      "grad_norm": 0.007140577305108309,
      "learning_rate": 0.00016,
      "loss": 12.4641,
      "step": 8
    },
    {
      "epoch": 0.005419238295950625,
      "grad_norm": 0.008437280543148518,
      "learning_rate": 0.00018,
      "loss": 12.4594,
      "step": 9
    },
    {
      "epoch": 0.005419238295950625,
      "eval_loss": 12.454813003540039,
      "eval_runtime": 6.8282,
      "eval_samples_per_second": 51.258,
      "eval_steps_per_second": 51.258,
      "step": 9
    },
    {
      "epoch": 0.006021375884389583,
      "grad_norm": 0.006291355937719345,
      "learning_rate": 0.0002,
      "loss": 12.442,
      "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": 6165626880.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}