{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0321455648218345,
  "eval_steps": 500,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12130401819560273,
      "grad_norm": 1.3961113691329956,
      "learning_rate": 7.960000000000001e-05,
      "loss": 2.5572,
      "mean_token_accuracy": 0.5595218382775784,
      "num_tokens": 57035.0,
      "step": 200
    },
    {
      "epoch": 0.24260803639120546,
      "grad_norm": 1.2175310850143433,
      "learning_rate": 9.686315789473685e-05,
      "loss": 1.2853,
      "mean_token_accuracy": 0.745141789726913,
      "num_tokens": 112951.0,
      "step": 400
    },
    {
      "epoch": 0.3639120545868082,
      "grad_norm": 1.0547959804534912,
      "learning_rate": 9.265263157894736e-05,
      "loss": 1.1237,
      "mean_token_accuracy": 0.773681266605854,
      "num_tokens": 169775.0,
      "step": 600
    },
    {
      "epoch": 0.4852160727824109,
      "grad_norm": 1.3259146213531494,
      "learning_rate": 8.844210526315791e-05,
      "loss": 0.9634,
      "mean_token_accuracy": 0.8042044347897171,
      "num_tokens": 226409.0,
      "step": 800
    },
    {
      "epoch": 0.6065200909780136,
      "grad_norm": 1.0659151077270508,
      "learning_rate": 8.423157894736843e-05,
      "loss": 0.8763,
      "mean_token_accuracy": 0.8207557284086943,
      "num_tokens": 283376.0,
      "step": 1000
    },
    {
      "epoch": 0.7278241091736164,
      "grad_norm": 0.9711537957191467,
      "learning_rate": 8.002105263157896e-05,
      "loss": 0.7978,
      "mean_token_accuracy": 0.8341990520060062,
      "num_tokens": 339940.0,
      "step": 1200
    },
    {
      "epoch": 0.849128127369219,
      "grad_norm": 0.7150431275367737,
      "learning_rate": 7.581052631578947e-05,
      "loss": 0.758,
      "mean_token_accuracy": 0.840606311634183,
      "num_tokens": 397189.0,
      "step": 1400
    },
    {
      "epoch": 0.9704321455648218,
      "grad_norm": 1.0585309267044067,
      "learning_rate": 7.16e-05,
      "loss": 0.6963,
      "mean_token_accuracy": 0.8530387983471155,
      "num_tokens": 453903.0,
      "step": 1600
    },
    {
      "epoch": 1.0915845337376802,
      "grad_norm": 0.7791032791137695,
      "learning_rate": 6.738947368421052e-05,
      "loss": 0.6373,
      "mean_token_accuracy": 0.8606961415019889,
      "num_tokens": 510914.0,
      "step": 1800
    },
    {
      "epoch": 1.2128885519332828,
      "grad_norm": 0.8289880752563477,
      "learning_rate": 6.317894736842105e-05,
      "loss": 0.6328,
      "mean_token_accuracy": 0.8607474016025662,
      "num_tokens": 567262.0,
      "step": 2000
    },
    {
      "epoch": 1.3341925701288855,
      "grad_norm": 0.9581061601638794,
      "learning_rate": 5.8968421052631585e-05,
      "loss": 0.5861,
      "mean_token_accuracy": 0.8699268980324268,
      "num_tokens": 622785.0,
      "step": 2200
    },
    {
      "epoch": 1.4554965883244884,
      "grad_norm": 0.7406579852104187,
      "learning_rate": 5.475789473684211e-05,
      "loss": 0.5775,
      "mean_token_accuracy": 0.8712259716168046,
      "num_tokens": 678802.0,
      "step": 2400
    },
    {
      "epoch": 1.576800606520091,
      "grad_norm": 0.9062987565994263,
      "learning_rate": 5.054736842105263e-05,
      "loss": 0.5801,
      "mean_token_accuracy": 0.8699949059635401,
      "num_tokens": 735965.0,
      "step": 2600
    },
    {
      "epoch": 1.6981046247156937,
      "grad_norm": 1.3214945793151855,
      "learning_rate": 4.6336842105263164e-05,
      "loss": 0.5514,
      "mean_token_accuracy": 0.875830891802907,
      "num_tokens": 793098.0,
      "step": 2800
    },
    {
      "epoch": 1.8194086429112963,
      "grad_norm": 0.99712735414505,
      "learning_rate": 4.212631578947369e-05,
      "loss": 0.5216,
      "mean_token_accuracy": 0.8817874767631292,
      "num_tokens": 849394.0,
      "step": 3000
    },
    {
      "epoch": 1.9407126611068992,
      "grad_norm": 0.968995213508606,
      "learning_rate": 3.791578947368421e-05,
      "loss": 0.5068,
      "mean_token_accuracy": 0.8846843484044075,
      "num_tokens": 906768.0,
      "step": 3200
    },
    {
      "epoch": 2.0618650492797572,
      "grad_norm": 0.8303135633468628,
      "learning_rate": 3.370526315789474e-05,
      "loss": 0.4856,
      "mean_token_accuracy": 0.8882180935748677,
      "num_tokens": 964437.0,
      "step": 3400
    },
    {
      "epoch": 2.1831690674753603,
      "grad_norm": 1.1362717151641846,
      "learning_rate": 2.9494736842105264e-05,
      "loss": 0.4485,
      "mean_token_accuracy": 0.8932469403743744,
      "num_tokens": 1021649.0,
      "step": 3600
    },
    {
      "epoch": 2.304473085670963,
      "grad_norm": 0.8629305362701416,
      "learning_rate": 2.528421052631579e-05,
      "loss": 0.4515,
      "mean_token_accuracy": 0.8930351422727107,
      "num_tokens": 1078206.0,
      "step": 3800
    },
    {
      "epoch": 2.4257771038665656,
      "grad_norm": 1.0377343893051147,
      "learning_rate": 2.1073684210526316e-05,
      "loss": 0.4472,
      "mean_token_accuracy": 0.8934314188361168,
      "num_tokens": 1135187.0,
      "step": 4000
    },
    {
      "epoch": 2.5470811220621683,
      "grad_norm": 1.061483383178711,
      "learning_rate": 1.686315789473684e-05,
      "loss": 0.4325,
      "mean_token_accuracy": 0.8970406632125377,
      "num_tokens": 1192547.0,
      "step": 4200
    },
    {
      "epoch": 2.668385140257771,
      "grad_norm": 0.8136045336723328,
      "learning_rate": 1.2652631578947368e-05,
      "loss": 0.4435,
      "mean_token_accuracy": 0.8965679884701967,
      "num_tokens": 1249416.0,
      "step": 4400
    },
    {
      "epoch": 2.7896891584533736,
      "grad_norm": 1.3490817546844482,
      "learning_rate": 8.463157894736843e-06,
      "loss": 0.4244,
      "mean_token_accuracy": 0.9009497262537479,
      "num_tokens": 1304893.0,
      "step": 4600
    },
    {
      "epoch": 2.9109931766489767,
      "grad_norm": 0.9819340109825134,
      "learning_rate": 4.252631578947369e-06,
      "loss": 0.4261,
      "mean_token_accuracy": 0.8986875934898854,
      "num_tokens": 1361211.0,
      "step": 4800
    },
    {
      "epoch": 3.0321455648218345,
      "grad_norm": 0.7312269806861877,
      "learning_rate": 4.210526315789474e-08,
      "loss": 0.4095,
      "mean_token_accuracy": 0.9022044740869047,
      "num_tokens": 1418483.0,
      "step": 5000
    }
  ],
  "logging_steps": 200,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 8666093203722240.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}