last-checkpoint/trainer_state.json

{
  "best_metric": 0.47778764367103577,
  "best_model_checkpoint": "miner_id_24/checkpoint-150",
  "epoch": 0.15873015873015872,
  "eval_steps": 50,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0010582010582010583,
      "eval_loss": 6.80908203125,
      "eval_runtime": 21.4913,
      "eval_samples_per_second": 18.519,
      "eval_steps_per_second": 4.653,
      "step": 1
    },
    {
      "epoch": 0.010582010582010581,
      "grad_norm": 10.603196144104004,
      "learning_rate": 5.05e-06,
      "loss": 4.4157,
      "step": 10
    },
    {
      "epoch": 0.021164021164021163,
      "grad_norm": 7.361603736877441,
      "learning_rate": 1.01e-05,
      "loss": 5.2851,
      "step": 20
    },
    {
      "epoch": 0.031746031746031744,
      "grad_norm": 7.728013038635254,
      "learning_rate": 9.323076923076924e-06,
      "loss": 4.741,
      "step": 30
    },
    {
      "epoch": 0.042328042328042326,
      "grad_norm": 8.335801124572754,
      "learning_rate": 8.546153846153847e-06,
      "loss": 4.5633,
      "step": 40
    },
    {
      "epoch": 0.05291005291005291,
      "grad_norm": 16.928085327148438,
      "learning_rate": 7.76923076923077e-06,
      "loss": 3.8241,
      "step": 50
    },
    {
      "epoch": 0.05291005291005291,
      "eval_loss": 2.496929883956909,
      "eval_runtime": 20.3952,
      "eval_samples_per_second": 19.514,
      "eval_steps_per_second": 4.903,
      "step": 50
    },
    {
      "epoch": 0.06349206349206349,
      "grad_norm": 8.544772148132324,
      "learning_rate": 6.992307692307692e-06,
      "loss": 1.2664,
      "step": 60
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 6.70974063873291,
      "learning_rate": 6.215384615384615e-06,
      "loss": 0.9053,
      "step": 70
    },
    {
      "epoch": 0.08465608465608465,
      "grad_norm": 7.614105224609375,
      "learning_rate": 5.438461538461538e-06,
      "loss": 0.7145,
      "step": 80
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 17.72666358947754,
      "learning_rate": 4.661538461538462e-06,
      "loss": 0.5158,
      "step": 90
    },
    {
      "epoch": 0.10582010582010581,
      "grad_norm": 8.977433204650879,
      "learning_rate": 3.884615384615385e-06,
      "loss": 0.8012,
      "step": 100
    },
    {
      "epoch": 0.10582010582010581,
      "eval_loss": 0.5890767574310303,
      "eval_runtime": 20.428,
      "eval_samples_per_second": 19.483,
      "eval_steps_per_second": 4.895,
      "step": 100
    },
    {
      "epoch": 0.1164021164021164,
      "grad_norm": 7.185388088226318,
      "learning_rate": 3.1076923076923076e-06,
      "loss": 0.3998,
      "step": 110
    },
    {
      "epoch": 0.12698412698412698,
      "grad_norm": 5.154370307922363,
      "learning_rate": 2.330769230769231e-06,
      "loss": 0.3106,
      "step": 120
    },
    {
      "epoch": 0.13756613756613756,
      "grad_norm": 11.753735542297363,
      "learning_rate": 1.5538461538461538e-06,
      "loss": 0.3945,
      "step": 130
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 8.9371976852417,
      "learning_rate": 7.769230769230769e-07,
      "loss": 0.438,
      "step": 140
    },
    {
      "epoch": 0.15873015873015872,
      "grad_norm": 16.33173942565918,
      "learning_rate": 0.0,
      "loss": 0.4991,
      "step": 150
    },
    {
      "epoch": 0.15873015873015872,
      "eval_loss": 0.47778764367103577,
      "eval_runtime": 20.4229,
      "eval_samples_per_second": 19.488,
      "eval_steps_per_second": 4.896,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.3320825700352e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}