last-checkpoint/trainer_state.json

{
  "best_metric": 0.8627253174781799,
  "best_model_checkpoint": "miner_id_24/checkpoint-288",
  "epoch": 3.096774193548387,
  "eval_steps": 96,
  "global_step": 576,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005376344086021506,
      "eval_loss": 2.6265158653259277,
      "eval_runtime": 6.7666,
      "eval_samples_per_second": 23.202,
      "eval_steps_per_second": 0.739,
      "step": 1
    },
    {
      "epoch": 0.5161290322580645,
      "grad_norm": 1.9268544912338257,
      "learning_rate": 0.000384,
      "loss": 2.8198,
      "step": 96
    },
    {
      "epoch": 0.5161290322580645,
      "eval_loss": 1.0718849897384644,
      "eval_runtime": 6.8186,
      "eval_samples_per_second": 23.025,
      "eval_steps_per_second": 0.733,
      "step": 96
    },
    {
      "epoch": 1.032258064516129,
      "grad_norm": 1.5161734819412231,
      "learning_rate": 0.00039997559251058097,
      "loss": 1.8322,
      "step": 192
    },
    {
      "epoch": 1.032258064516129,
      "eval_loss": 0.9115006923675537,
      "eval_runtime": 6.8277,
      "eval_samples_per_second": 22.995,
      "eval_steps_per_second": 0.732,
      "step": 192
    },
    {
      "epoch": 1.5483870967741935,
      "grad_norm": 1.7862235307693481,
      "learning_rate": 0.00039989808570615163,
      "loss": 1.4055,
      "step": 288
    },
    {
      "epoch": 1.5483870967741935,
      "eval_loss": 0.8627253174781799,
      "eval_runtime": 6.8612,
      "eval_samples_per_second": 22.882,
      "eval_steps_per_second": 0.729,
      "step": 288
    },
    {
      "epoch": 2.064516129032258,
      "grad_norm": 1.742639422416687,
      "learning_rate": 0.0003997674540450395,
      "loss": 1.2677,
      "step": 384
    },
    {
      "epoch": 2.064516129032258,
      "eval_loss": 0.8888482451438904,
      "eval_runtime": 6.8571,
      "eval_samples_per_second": 22.896,
      "eval_steps_per_second": 0.729,
      "step": 384
    },
    {
      "epoch": 2.5806451612903225,
      "grad_norm": 2.217721939086914,
      "learning_rate": 0.00039958373224387655,
      "loss": 0.9139,
      "step": 480
    },
    {
      "epoch": 2.5806451612903225,
      "eval_loss": 0.8767105340957642,
      "eval_runtime": 6.9026,
      "eval_samples_per_second": 22.745,
      "eval_steps_per_second": 0.724,
      "step": 480
    },
    {
      "epoch": 3.096774193548387,
      "grad_norm": 2.5925180912017822,
      "learning_rate": 0.0003993469691285149,
      "loss": 0.9038,
      "step": 576
    },
    {
      "epoch": 3.096774193548387,
      "eval_loss": 0.99139404296875,
      "eval_runtime": 7.0307,
      "eval_samples_per_second": 22.331,
      "eval_steps_per_second": 0.711,
      "step": 576
    }
  ],
  "logging_steps": 96,
  "max_steps": 18600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 96,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8750975648989184e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}