last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0002880785878387624,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.880785878387624e-05,
      "grad_norm": 3.4727489948272705,
      "learning_rate": 2e-05,
      "loss": 3.7208,
      "step": 1
    },
    {
      "epoch": 2.880785878387624e-05,
      "eval_loss": 4.868256568908691,
      "eval_runtime": 445.498,
      "eval_samples_per_second": 32.808,
      "eval_steps_per_second": 16.404,
      "step": 1
    },
    {
      "epoch": 5.761571756775248e-05,
      "grad_norm": 3.298544406890869,
      "learning_rate": 4e-05,
      "loss": 3.9491,
      "step": 2
    },
    {
      "epoch": 8.642357635162872e-05,
      "grad_norm": 5.727396011352539,
      "learning_rate": 6e-05,
      "loss": 4.5353,
      "step": 3
    },
    {
      "epoch": 8.642357635162872e-05,
      "eval_loss": 4.863677501678467,
      "eval_runtime": 444.524,
      "eval_samples_per_second": 32.88,
      "eval_steps_per_second": 16.44,
      "step": 3
    },
    {
      "epoch": 0.00011523143513550496,
      "grad_norm": 6.3204874992370605,
      "learning_rate": 8e-05,
      "loss": 4.7909,
      "step": 4
    },
    {
      "epoch": 0.0001440392939193812,
      "grad_norm": 3.4016435146331787,
      "learning_rate": 0.0001,
      "loss": 4.6886,
      "step": 5
    },
    {
      "epoch": 0.00017284715270325744,
      "grad_norm": 3.6947529315948486,
      "learning_rate": 0.00012,
      "loss": 5.0971,
      "step": 6
    },
    {
      "epoch": 0.00017284715270325744,
      "eval_loss": 4.807370185852051,
      "eval_runtime": 444.517,
      "eval_samples_per_second": 32.881,
      "eval_steps_per_second": 16.44,
      "step": 6
    },
    {
      "epoch": 0.0002016550114871337,
      "grad_norm": 4.251638412475586,
      "learning_rate": 0.00014,
      "loss": 4.2869,
      "step": 7
    },
    {
      "epoch": 0.00023046287027100992,
      "grad_norm": 4.2845869064331055,
      "learning_rate": 0.00016,
      "loss": 4.553,
      "step": 8
    },
    {
      "epoch": 0.00025927072905488615,
      "grad_norm": 4.339658260345459,
      "learning_rate": 0.00018,
      "loss": 3.9509,
      "step": 9
    },
    {
      "epoch": 0.00025927072905488615,
      "eval_loss": 4.4670329093933105,
      "eval_runtime": 445.395,
      "eval_samples_per_second": 32.816,
      "eval_steps_per_second": 16.408,
      "step": 9
    },
    {
      "epoch": 0.0002880785878387624,
      "grad_norm": 6.038758754730225,
      "learning_rate": 0.0002,
      "loss": 4.4828,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}