trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 208,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 0.04972561076283455,
      "learning_rate": 0.000509291899366086,
      "loss": 0.2396,
      "mean_token_accuracy": 0.9078072667121887,
      "num_tokens": 10242232.0,
      "step": 25
    },
    {
      "epoch": 0.963855421686747,
      "grad_norm": 0.04746510088443756,
      "learning_rate": 0.0004965277770447238,
      "loss": 0.1033,
      "mean_token_accuracy": 0.9155593666434289,
      "num_tokens": 20493886.0,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.9735881001391309,
      "eval_num_tokens": 21054178.0,
      "eval_runtime": 41.8061,
      "eval_samples_per_second": 8.826,
      "eval_steps_per_second": 1.124,
      "step": 52
    },
    {
      "epoch": 1.4433734939759035,
      "grad_norm": 0.048190850764513016,
      "learning_rate": 0.00047188122815287187,
      "loss": 0.0796,
      "mean_token_accuracy": 0.9210376496890083,
      "num_tokens": 30877786.0,
      "step": 75
    },
    {
      "epoch": 1.9253012048192772,
      "grad_norm": 0.07209828495979309,
      "learning_rate": 0.0004365673027192623,
      "loss": 0.063,
      "mean_token_accuracy": 0.9394274836778641,
      "num_tokens": 41107523.0,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.9802652305745064,
      "eval_num_tokens": 42118832.0,
      "eval_runtime": 43.7261,
      "eval_samples_per_second": 8.439,
      "eval_steps_per_second": 1.075,
      "step": 104
    },
    {
      "epoch": 2.404819277108434,
      "grad_norm": 0.05603017657995224,
      "learning_rate": 0.00039232694168865086,
      "loss": 0.0545,
      "mean_token_accuracy": 0.9359195819452181,
      "num_tokens": 51238167.0,
      "step": 125
    },
    {
      "epoch": 2.886746987951807,
      "grad_norm": 0.07570821046829224,
      "learning_rate": 0.00034134115028725524,
      "loss": 0.0391,
      "mean_token_accuracy": 0.9296775516867638,
      "num_tokens": 61475716.0,
      "step": 150
    },
    {
      "epoch": 3.0,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.9825496534083752,
      "eval_num_tokens": 63186871.0,
      "eval_runtime": 42.5381,
      "eval_samples_per_second": 8.675,
      "eval_steps_per_second": 1.105,
      "step": 156
    },
    {
      "epoch": 3.3662650602409636,
      "grad_norm": 0.07343257963657379,
      "learning_rate": 0.0002861234766624867,
      "loss": 0.0345,
      "mean_token_accuracy": 0.9456785363168573,
      "num_tokens": 71027426.0,
      "step": 175
    },
    {
      "epoch": 3.8481927710843373,
      "grad_norm": 0.04622579738497734,
      "learning_rate": 0.0002293960964917063,
      "loss": 0.0259,
      "mean_token_accuracy": 0.9330729904770851,
      "num_tokens": 81341930.0,
      "step": 200
    },
    {
      "epoch": 4.0,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.9852228583173549,
      "eval_num_tokens": 84250831.0,
      "eval_runtime": 41.6702,
      "eval_samples_per_second": 8.855,
      "eval_steps_per_second": 1.128,
      "step": 208
    }
  ],
  "logging_steps": 25,
  "max_steps": 364,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.6994667840222003e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}