trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.909090909090909,
  "eval_steps": 500,
  "global_step": 72,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04040404040404041,
      "grad_norm": 2.354057369065775,
      "learning_rate": 2.5e-05,
      "loss": 3.6821,
      "step": 1
    },
    {
      "epoch": 0.20202020202020202,
      "grad_norm": 3.318817777883566,
      "learning_rate": 0.000125,
      "loss": 3.6411,
      "step": 5
    },
    {
      "epoch": 0.40404040404040403,
      "grad_norm": 1.1567295340838406,
      "learning_rate": 0.0001995184726672197,
      "loss": 2.3671,
      "step": 10
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 0.5651519215666836,
      "learning_rate": 0.00019415440651830208,
      "loss": 1.478,
      "step": 15
    },
    {
      "epoch": 0.8080808080808081,
      "grad_norm": 0.4028700304940561,
      "learning_rate": 0.00018314696123025454,
      "loss": 1.3252,
      "step": 20
    },
    {
      "epoch": 0.9696969696969697,
      "eval_loss": 1.1692708730697632,
      "eval_runtime": 177.3352,
      "eval_samples_per_second": 35.644,
      "eval_steps_per_second": 2.233,
      "step": 24
    },
    {
      "epoch": 1.0101010101010102,
      "grad_norm": 0.3079834505872443,
      "learning_rate": 0.00016715589548470185,
      "loss": 1.2706,
      "step": 25
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 0.25648814038671125,
      "learning_rate": 0.0001471396736825998,
      "loss": 1.1881,
      "step": 30
    },
    {
      "epoch": 1.4141414141414141,
      "grad_norm": 0.2008341133631863,
      "learning_rate": 0.0001242980179903264,
      "loss": 1.1514,
      "step": 35
    },
    {
      "epoch": 1.6161616161616161,
      "grad_norm": 0.1814019139468598,
      "learning_rate": 0.0001,
      "loss": 1.1815,
      "step": 40
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.18331867012606626,
      "learning_rate": 7.570198200967362e-05,
      "loss": 1.1352,
      "step": 45
    },
    {
      "epoch": 1.9797979797979797,
      "eval_loss": 1.0709240436553955,
      "eval_runtime": 175.2955,
      "eval_samples_per_second": 36.059,
      "eval_steps_per_second": 2.259,
      "step": 49
    },
    {
      "epoch": 2.0202020202020203,
      "grad_norm": 0.2143290425477542,
      "learning_rate": 5.286032631740023e-05,
      "loss": 1.1283,
      "step": 50
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.18532120253500695,
      "learning_rate": 3.2844104515298155e-05,
      "loss": 1.112,
      "step": 55
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 0.19523394937897012,
      "learning_rate": 1.6853038769745467e-05,
      "loss": 1.1094,
      "step": 60
    },
    {
      "epoch": 2.6262626262626263,
      "grad_norm": 0.19014746927385398,
      "learning_rate": 5.8455934816979305e-06,
      "loss": 1.1021,
      "step": 65
    },
    {
      "epoch": 2.8282828282828283,
      "grad_norm": 0.1651175400452345,
      "learning_rate": 4.815273327803182e-07,
      "loss": 1.1558,
      "step": 70
    },
    {
      "epoch": 2.909090909090909,
      "step": 72,
      "total_flos": 644347544469504.0,
      "train_loss": 0.0,
      "train_runtime": 0.0104,
      "train_samples_per_second": 1824024.464,
      "train_steps_per_second": 6925.579
    }
  ],
  "logging_steps": 5,
  "max_steps": 72,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 644347544469504.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}