{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.4615384615384617,
  "eval_steps": 3,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_loss": 2.2738115787506104,
      "eval_runtime": 374.5242,
      "eval_samples_per_second": 0.635,
      "eval_steps_per_second": 0.318,
      "memory/device_mem_reserved(gib)": 31.86,
      "memory/max_mem_active(gib)": 31.55,
      "memory/max_mem_allocated(gib)": 31.55,
      "step": 0
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.4899442195892334,
      "learning_rate": 0.0,
      "loss": 2.2369,
      "memory/device_mem_reserved(gib)": 38.32,
      "memory/max_mem_active(gib)": 34.44,
      "memory/max_mem_allocated(gib)": 34.44,
      "step": 1
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.5000583529472351,
      "learning_rate": 0.00010000000474974513,
      "loss": 2.2373,
      "memory/device_mem_reserved(gib)": 38.54,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 2
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.39517298340797424,
      "learning_rate": 0.00020000000949949026,
      "loss": 2.2188,
      "memory/device_mem_reserved(gib)": 38.54,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 3
    },
    {
      "epoch": 0.46153846153846156,
      "eval_loss": 2.105576515197754,
      "eval_runtime": 376.3787,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.54,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 3
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.12939366698265076,
      "learning_rate": 0.0003000000142492354,
      "loss": 2.1136,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 4
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.12260633707046509,
      "learning_rate": 0.0002989857748616487,
      "loss": 2.0954,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 5
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.15485626459121704,
      "learning_rate": 0.0002959567354992032,
      "loss": 2.0845,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 6
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 2.0672547817230225,
      "eval_runtime": 376.5486,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 6
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.12021517008543015,
      "learning_rate": 0.0002909539034590125,
      "loss": 2.077,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 7
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 0.08224741369485855,
      "learning_rate": 0.00028404491604305804,
      "loss": 2.0345,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 8
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.06893477588891983,
      "learning_rate": 0.0002753231965471059,
      "loss": 2.0332,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 9
    },
    {
      "epoch": 1.3076923076923077,
      "eval_loss": 2.044691801071167,
      "eval_runtime": 376.6259,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 9
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 0.08403533697128296,
      "learning_rate": 0.0002649066736921668,
      "loss": 2.0445,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 10
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.08370130509138107,
      "learning_rate": 0.00025293626822531223,
      "loss": 2.0298,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 11
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 0.07989141345024109,
      "learning_rate": 0.00023957379744388163,
      "loss": 2.0284,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 12
    },
    {
      "epoch": 1.7692307692307692,
      "eval_loss": 2.0312037467956543,
      "eval_runtime": 376.3006,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 12
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.08033885806798935,
      "learning_rate": 0.00022500001068692654,
      "loss": 2.0108,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 13
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.10752119868993759,
      "learning_rate": 0.00020941196999046952,
      "loss": 1.9729,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 14
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.08343259990215302,
      "learning_rate": 0.0001930204889504239,
      "loss": 1.9655,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 15
    },
    {
      "epoch": 2.1538461538461537,
      "eval_loss": 2.0194833278656006,
      "eval_runtime": 376.3527,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 15
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.08610356599092484,
      "learning_rate": 0.00017604722233954817,
      "loss": 1.9656,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 16
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.08564396947622299,
      "learning_rate": 0.00015872172662056983,
      "loss": 1.976,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 17
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.08958195149898529,
      "learning_rate": 0.00014127828762866557,
      "loss": 1.981,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 18
    },
    {
      "epoch": 2.6153846153846154,
      "eval_loss": 2.014101505279541,
      "eval_runtime": 376.448,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 18
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.10577207803726196,
      "learning_rate": 0.000123952777357772,
      "loss": 1.961,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 19
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.09847240895032883,
      "learning_rate": 0.00010697951802285388,
      "loss": 1.9455,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1634841412305832,
      "learning_rate": 9.058803698280826e-05,
      "loss": 1.9211,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 21
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.0099527835845947,
      "eval_runtime": 376.3789,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 21
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 0.10216221213340759,
      "learning_rate": 7.500000356230885e-05,
      "loss": 1.9455,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 22
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.11089015752077103,
      "learning_rate": 6.042621680535376e-05,
      "loss": 1.9001,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 23
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 0.10617231577634811,
      "learning_rate": 4.706375693785958e-05,
      "loss": 1.9149,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 24
    },
    {
      "epoch": 3.4615384615384617,
      "eval_loss": 2.009589910507202,
      "eval_runtime": 376.3891,
      "eval_samples_per_second": 0.632,
      "eval_steps_per_second": 0.316,
      "memory/device_mem_reserved(gib)": 38.55,
      "memory/max_mem_active(gib)": 34.71,
      "memory/max_mem_allocated(gib)": 34.71,
      "step": 24
    }
  ],
  "logging_steps": 1,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 6,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.829824816012329e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}