{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6736842105263157,
  "eval_steps": 12,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_loss": 1.2786176204681396,
      "eval_ppl": 3.59167,
      "eval_runtime": 276.2614,
      "eval_samples_per_second": 0.576,
      "eval_steps_per_second": 0.576,
      "memory/device_reserved (GiB)": 42.45,
      "memory/max_active (GiB)": 40.26,
      "memory/max_allocated (GiB)": 38.06,
      "step": 0
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 4.365630626678467,
      "learning_rate": 7e-06,
      "loss": 1.1477760314941405,
      "memory/device_reserved (GiB)": 71.27,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 3.15118,
      "step": 5,
      "tokens/total": 2621440,
      "tokens/trainable": 489230
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 2.4648349285125732,
      "learning_rate": 6.952144421988455e-06,
      "loss": 0.9022371292114257,
      "memory/device_reserved (GiB)": 71.27,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 2.46511,
      "step": 10,
      "tokens/total": 5242880,
      "tokens/train_per_sec_per_gpu": 225.38,
      "tokens/trainable": 896733
    },
    {
      "epoch": 0.25263157894736843,
      "eval_loss": 0.7994529604911804,
      "eval_ppl": 2.22432,
      "eval_runtime": 276.9872,
      "eval_samples_per_second": 0.574,
      "eval_steps_per_second": 0.574,
      "memory/device_reserved (GiB)": 71.27,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "step": 12
    },
    {
      "epoch": 0.3157894736842105,
      "grad_norm": 1.6509625911712646,
      "learning_rate": 6.810031755475611e-06,
      "loss": 0.7818592548370361,
      "memory/device_reserved (GiB)": 74.72,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 2.18553,
      "step": 15,
      "tokens/total": 7864320,
      "tokens/train_per_sec_per_gpu": 208.77,
      "tokens/trainable": 1345756
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 1.7147045135498047,
      "learning_rate": 6.577980021920982e-06,
      "loss": 0.7099627494812012,
      "memory/device_reserved (GiB)": 74.72,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 2.03392,
      "step": 20,
      "tokens/total": 10485760,
      "tokens/train_per_sec_per_gpu": 361.48,
      "tokens/trainable": 1808219
    },
    {
      "epoch": 0.5052631578947369,
      "eval_loss": 0.6555427312850952,
      "eval_ppl": 1.92619,
      "eval_runtime": 248.0295,
      "eval_samples_per_second": 0.641,
      "eval_steps_per_second": 0.641,
      "memory/device_reserved (GiB)": 74.72,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "step": 24
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 1.5091627836227417,
      "learning_rate": 6.26303999582478e-06,
      "loss": 0.6799451828002929,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.97377,
      "step": 25,
      "tokens/total": 13107200,
      "tokens/train_per_sec_per_gpu": 268.83,
      "tokens/trainable": 2291286
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": 1.6102690696716309,
      "learning_rate": 5.874780970512599e-06,
      "loss": 0.625995397567749,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.87011,
      "step": 30,
      "tokens/total": 15728640,
      "tokens/train_per_sec_per_gpu": 380.95,
      "tokens/trainable": 2755207
    },
    {
      "epoch": 0.7368421052631579,
      "grad_norm": 1.7557069063186646,
      "learning_rate": 5.425e-06,
      "loss": 0.5982392787933349,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.81891,
      "step": 35,
      "tokens/total": 18350080,
      "tokens/train_per_sec_per_gpu": 309.14,
      "tokens/trainable": 3213136
    },
    {
      "epoch": 0.7578947368421053,
      "eval_loss": 0.5690193176269531,
      "eval_ppl": 1.76653,
      "eval_runtime": 247.7474,
      "eval_samples_per_second": 0.642,
      "eval_steps_per_second": 0.642,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "step": 36
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 1.497903823852539,
      "learning_rate": 4.927363451475857e-06,
      "loss": 0.5714361667633057,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.77081,
      "step": 40,
      "tokens/total": 20971520,
      "tokens/train_per_sec_per_gpu": 325.41,
      "tokens/trainable": 3696644
    },
    {
      "epoch": 0.9473684210526315,
      "grad_norm": 1.6041510105133057,
      "learning_rate": 4.396991759650831e-06,
      "loss": 0.5171830177307128,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.6773,
      "step": 45,
      "tokens/total": 23592960,
      "tokens/train_per_sec_per_gpu": 298.87,
      "tokens/trainable": 4143446
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5053056478500366,
      "eval_ppl": 1.65749,
      "eval_runtime": 247.076,
      "eval_samples_per_second": 0.644,
      "eval_steps_per_second": 0.644,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "step": 48
    },
    {
      "epoch": 1.0421052631578946,
      "grad_norm": 1.7307233810424805,
      "learning_rate": 3.85e-06,
      "loss": 0.47033162117004396,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.60052,
      "step": 50,
      "tokens/total": 25952256,
      "tokens/train_per_sec_per_gpu": 237.38,
      "tokens/trainable": 4500374
    },
    {
      "epoch": 1.1473684210526316,
      "grad_norm": 1.4362328052520752,
      "learning_rate": 3.3030082403491696e-06,
      "loss": 0.43146610260009766,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.53951,
      "step": 55,
      "tokens/total": 28573696,
      "tokens/train_per_sec_per_gpu": 266.77,
      "tokens/trainable": 4937656
    },
    {
      "epoch": 1.2526315789473683,
      "grad_norm": 1.665703535079956,
      "learning_rate": 2.7726365485241435e-06,
      "loss": 0.4059854030609131,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.50078,
      "step": 60,
      "tokens/total": 31195136,
      "tokens/train_per_sec_per_gpu": 312.86,
      "tokens/trainable": 5365027
    },
    {
      "epoch": 1.2526315789473683,
      "eval_loss": 0.46400853991508484,
      "eval_ppl": 1.59044,
      "eval_runtime": 245.9575,
      "eval_samples_per_second": 0.646,
      "eval_steps_per_second": 0.646,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 40.28,
      "memory/max_allocated (GiB)": 38.08,
      "step": 60
    },
    {
      "epoch": 1.3578947368421053,
      "grad_norm": 1.6653640270233154,
      "learning_rate": 2.2750000000000007e-06,
      "loss": 0.3871333599090576,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.47275,
      "step": 65,
      "tokens/total": 33816576,
      "tokens/train_per_sec_per_gpu": 235.21,
      "tokens/trainable": 5817060
    },
    {
      "epoch": 1.4631578947368422,
      "grad_norm": 1.733436942100525,
      "learning_rate": 1.8252190294874008e-06,
      "loss": 0.38542895317077636,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.47024,
      "step": 70,
      "tokens/total": 36438016,
      "tokens/train_per_sec_per_gpu": 326.88,
      "tokens/trainable": 6266148
    },
    {
      "epoch": 1.5052631578947369,
      "eval_loss": 0.43543583154678345,
      "eval_ppl": 1.54564,
      "eval_runtime": 250.6228,
      "eval_samples_per_second": 0.634,
      "eval_steps_per_second": 0.634,
      "memory/device_reserved (GiB)": 71.11,
      "memory/max_active (GiB)": 53.96,
      "memory/max_allocated (GiB)": 51.79,
      "step": 72
    },
    {
      "epoch": 1.568421052631579,
      "grad_norm": 1.515413522720337,
      "learning_rate": 1.4369600041752197e-06,
      "loss": 0.3665297508239746,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.44272,
      "step": 75,
      "tokens/total": 39059456,
      "tokens/train_per_sec_per_gpu": 239.62,
      "tokens/trainable": 6725806
    },
    {
      "epoch": 1.6736842105263157,
      "grad_norm": 1.5554890632629395,
      "learning_rate": 1.122019978079018e-06,
      "loss": 0.34335381984710694,
      "memory/device_reserved (GiB)": 74.7,
      "memory/max_active (GiB)": 53.99,
      "memory/max_allocated (GiB)": 51.79,
      "ppl": 1.40967,
      "step": 80,
      "tokens/total": 41680896,
      "tokens/train_per_sec_per_gpu": 344.08,
      "tokens/trainable": 7193319
    }
  ],
  "logging_steps": 5,
  "max_steps": 94,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 366247445266432.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}