trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 7048,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14188422247446084,
      "grad_norm": 24.020404815673828,
      "learning_rate": 2.787173666288309e-05,
      "loss": 1.2482,
      "step": 500
    },
    {
      "epoch": 0.28376844494892167,
      "grad_norm": 15.585792541503906,
      "learning_rate": 2.5743473325766176e-05,
      "loss": 0.921,
      "step": 1000
    },
    {
      "epoch": 0.42565266742338254,
      "grad_norm": 15.092517852783203,
      "learning_rate": 2.3615209988649265e-05,
      "loss": 0.8583,
      "step": 1500
    },
    {
      "epoch": 0.5675368898978433,
      "grad_norm": 12.11979866027832,
      "learning_rate": 2.1486946651532347e-05,
      "loss": 0.8274,
      "step": 2000
    },
    {
      "epoch": 0.7094211123723042,
      "grad_norm": 18.533437728881836,
      "learning_rate": 1.9358683314415437e-05,
      "loss": 0.7728,
      "step": 2500
    },
    {
      "epoch": 0.8513053348467651,
      "grad_norm": 16.716705322265625,
      "learning_rate": 1.7230419977298526e-05,
      "loss": 0.7614,
      "step": 3000
    },
    {
      "epoch": 0.9931895573212258,
      "grad_norm": 18.170560836791992,
      "learning_rate": 1.5102156640181612e-05,
      "loss": 0.7471,
      "step": 3500
    },
    {
      "epoch": 1.1350737797956867,
      "grad_norm": 11.996767044067383,
      "learning_rate": 1.29738933030647e-05,
      "loss": 0.6015,
      "step": 4000
    },
    {
      "epoch": 1.2769580022701477,
      "grad_norm": 15.390630722045898,
      "learning_rate": 1.0845629965947787e-05,
      "loss": 0.6034,
      "step": 4500
    },
    {
      "epoch": 1.4188422247446084,
      "grad_norm": 14.796677589416504,
      "learning_rate": 8.717366628830874e-06,
      "loss": 0.5894,
      "step": 5000
    },
    {
      "epoch": 1.5607264472190692,
      "grad_norm": 11.31975269317627,
      "learning_rate": 6.589103291713962e-06,
      "loss": 0.5715,
      "step": 5500
    },
    {
      "epoch": 1.70261066969353,
      "grad_norm": 13.917352676391602,
      "learning_rate": 4.4608399545970485e-06,
      "loss": 0.5731,
      "step": 6000
    },
    {
      "epoch": 1.844494892167991,
      "grad_norm": 16.910314559936523,
      "learning_rate": 2.332576617480136e-06,
      "loss": 0.5691,
      "step": 6500
    },
    {
      "epoch": 1.9863791146424519,
      "grad_norm": 14.70508098602295,
      "learning_rate": 2.043132803632236e-07,
      "loss": 0.5752,
      "step": 7000
    },
    {
      "epoch": 2.0,
      "step": 7048,
      "total_flos": 5601854796788736.0,
      "train_loss": 0.7287338001909375,
      "train_runtime": 2646.7279,
      "train_samples_per_second": 127.787,
      "train_steps_per_second": 2.663
    }
  ],
  "logging_steps": 500,
  "max_steps": 7048,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5601854796788736.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": null
}