trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9978143668949438,
  "eval_steps": 22,
  "global_step": 428,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0512895235319831,
      "grad_norm": 1.549578309059143,
      "learning_rate": 0.0015811388300841895,
      "loss": 1.5548,
      "step": 22
    },
    {
      "epoch": 0.1025790470639662,
      "grad_norm": 0.3865428864955902,
      "learning_rate": 0.0008304547985373997,
      "loss": 1.3048,
      "step": 44
    },
    {
      "epoch": 0.1538685705959493,
      "grad_norm": 0.4011533260345459,
      "learning_rate": 0.0006262242910851496,
      "loss": 1.1538,
      "step": 66
    },
    {
      "epoch": 0.2051580941279324,
      "grad_norm": 0.4274662435054779,
      "learning_rate": 0.0005234239225902136,
      "loss": 1.1093,
      "step": 88
    },
    {
      "epoch": 0.25644761765991547,
      "grad_norm": 0.33565089106559753,
      "learning_rate": 0.00046126560401444256,
      "loss": 1.0236,
      "step": 110
    },
    {
      "epoch": 0.3077371411918986,
      "grad_norm": 0.39469897747039795,
      "learning_rate": 0.00041522739926869986,
      "loss": 1.0122,
      "step": 132
    },
    {
      "epoch": 0.35902666472388167,
      "grad_norm": 0.4309040307998657,
      "learning_rate": 0.0003806934938134405,
      "loss": 0.9694,
      "step": 154
    },
    {
      "epoch": 0.4103161882558648,
      "grad_norm": 0.4251372218132019,
      "learning_rate": 0.00035355339059327376,
      "loss": 0.9781,
      "step": 176
    },
    {
      "epoch": 0.46160571178784787,
      "grad_norm": 0.3717755079269409,
      "learning_rate": 0.00033149677206589795,
      "loss": 0.9694,
      "step": 198
    },
    {
      "epoch": 0.5128952353198309,
      "grad_norm": 0.39009973406791687,
      "learning_rate": 0.0003131121455425748,
      "loss": 0.9111,
      "step": 220
    },
    {
      "epoch": 0.5641847588518141,
      "grad_norm": 0.3686709403991699,
      "learning_rate": 0.0002974820586543648,
      "loss": 0.9077,
      "step": 242
    },
    {
      "epoch": 0.6154742823837972,
      "grad_norm": 0.4044100344181061,
      "learning_rate": 0.0002839809171235324,
      "loss": 0.8875,
      "step": 264
    },
    {
      "epoch": 0.6667638059157802,
      "grad_norm": 0.3807663917541504,
      "learning_rate": 0.0002721655269759087,
      "loss": 0.8618,
      "step": 286
    },
    {
      "epoch": 0.7180533294477633,
      "grad_norm": 0.4056909680366516,
      "learning_rate": 0.0002617119612951068,
      "loss": 0.8406,
      "step": 308
    },
    {
      "epoch": 0.7693428529797465,
      "grad_norm": 0.3789248466491699,
      "learning_rate": 0.00025237723256253436,
      "loss": 0.8282,
      "step": 330
    },
    {
      "epoch": 0.8206323765117296,
      "grad_norm": 0.3973967134952545,
      "learning_rate": 0.00024397501823713327,
      "loss": 0.8097,
      "step": 352
    },
    {
      "epoch": 0.8719219000437126,
      "grad_norm": 0.4864844083786011,
      "learning_rate": 0.00023635972962353274,
      "loss": 0.7718,
      "step": 374
    },
    {
      "epoch": 0.9232114235756957,
      "grad_norm": 0.571861207485199,
      "learning_rate": 0.00022941573387056174,
      "loss": 0.77,
      "step": 396
    },
    {
      "epoch": 0.9745009471076789,
      "grad_norm": 0.3887988328933716,
      "learning_rate": 0.00022304986837273525,
      "loss": 0.7795,
      "step": 418
    },
    {
      "epoch": 0.9978143668949438,
      "step": 428,
      "total_flos": 4.3469664994124104e+18,
      "train_loss": 0.9656199194560541,
      "train_runtime": 3172.7582,
      "train_samples_per_second": 17.303,
      "train_steps_per_second": 0.135
    }
  ],
  "logging_steps": 22,
  "max_steps": 428,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 22,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.3469664994124104e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}