{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999050933248972,
  "eval_steps": 1000,
  "global_step": 4740,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06327111673521038,
      "grad_norm": 4.5707688331604,
      "learning_rate": 1.0548523206751056e-05,
      "loss": 6.0536,
      "step": 100
    },
    {
      "epoch": 0.12654223347042076,
      "grad_norm": 7.803924083709717,
      "learning_rate": 2.1097046413502112e-05,
      "loss": 5.8177,
      "step": 200
    },
    {
      "epoch": 0.18981335020563114,
      "grad_norm": 8.050126075744629,
      "learning_rate": 3.1645569620253167e-05,
      "loss": 5.0311,
      "step": 300
    },
    {
      "epoch": 0.2530844669408415,
      "grad_norm": 8.803975105285645,
      "learning_rate": 4.2194092827004224e-05,
      "loss": 4.7351,
      "step": 400
    },
    {
      "epoch": 0.3163555836760519,
      "grad_norm": 12.846040725708008,
      "learning_rate": 4.96952648851383e-05,
      "loss": 4.5453,
      "step": 500
    },
    {
      "epoch": 0.3796267004112623,
      "grad_norm": 15.128095626831055,
      "learning_rate": 4.852320675105486e-05,
      "loss": 4.3979,
      "step": 600
    },
    {
      "epoch": 0.44289781714647264,
      "grad_norm": 10.33956527709961,
      "learning_rate": 4.7351148616971405e-05,
      "loss": 4.4056,
      "step": 700
    },
    {
      "epoch": 0.506168933881683,
      "grad_norm": 23.287179946899414,
      "learning_rate": 4.617909048288795e-05,
      "loss": 4.2787,
      "step": 800
    },
    {
      "epoch": 0.5694400506168934,
      "grad_norm": 12.955565452575684,
      "learning_rate": 4.50070323488045e-05,
      "loss": 4.3504,
      "step": 900
    },
    {
      "epoch": 0.6327111673521038,
      "grad_norm": 10.07479476928711,
      "learning_rate": 4.3834974214721055e-05,
      "loss": 4.2521,
      "step": 1000
    },
    {
      "epoch": 0.6327111673521038,
      "eval_runtime": 33.0567,
      "eval_samples_per_second": 95.593,
      "eval_steps_per_second": 11.949,
      "step": 1000
    },
    {
      "epoch": 0.6959822840873141,
      "grad_norm": 12.32780647277832,
      "learning_rate": 4.26629160806376e-05,
      "loss": 4.2314,
      "step": 1100
    },
    {
      "epoch": 0.7592534008225246,
      "grad_norm": 15.515801429748535,
      "learning_rate": 4.149085794655415e-05,
      "loss": 4.3058,
      "step": 1200
    },
    {
      "epoch": 0.8225245175577349,
      "grad_norm": 12.472834587097168,
      "learning_rate": 4.03187998124707e-05,
      "loss": 4.2158,
      "step": 1300
    },
    {
      "epoch": 0.8857956342929453,
      "grad_norm": 16.903112411499023,
      "learning_rate": 3.914674167838725e-05,
      "loss": 4.142,
      "step": 1400
    },
    {
      "epoch": 0.9490667510281556,
      "grad_norm": 13.487791061401367,
      "learning_rate": 3.79746835443038e-05,
      "loss": 4.0534,
      "step": 1500
    },
    {
      "epoch": 1.012337867763366,
      "grad_norm": 14.721494674682617,
      "learning_rate": 3.680262541022035e-05,
      "loss": 4.1405,
      "step": 1600
    },
    {
      "epoch": 1.0756089844985763,
      "grad_norm": 16.011690139770508,
      "learning_rate": 3.56305672761369e-05,
      "loss": 4.1192,
      "step": 1700
    },
    {
      "epoch": 1.1388801012337868,
      "grad_norm": 15.692846298217773,
      "learning_rate": 3.445850914205345e-05,
      "loss": 4.1407,
      "step": 1800
    },
    {
      "epoch": 1.2021512179689973,
      "grad_norm": 13.71811294555664,
      "learning_rate": 3.328645100797e-05,
      "loss": 4.1839,
      "step": 1900
    },
    {
      "epoch": 1.2654223347042075,
      "grad_norm": 13.474448204040527,
      "learning_rate": 3.2114392873886545e-05,
      "loss": 4.0813,
      "step": 2000
    },
    {
      "epoch": 1.2654223347042075,
      "eval_runtime": 33.0889,
      "eval_samples_per_second": 95.5,
      "eval_steps_per_second": 11.938,
      "step": 2000
    },
    {
      "epoch": 1.328693451439418,
      "grad_norm": 15.276654243469238,
      "learning_rate": 3.09423347398031e-05,
      "loss": 3.9624,
      "step": 2100
    },
    {
      "epoch": 1.3919645681746282,
      "grad_norm": 13.796238899230957,
      "learning_rate": 2.9770276605719643e-05,
      "loss": 4.0653,
      "step": 2200
    },
    {
      "epoch": 1.4552356849098387,
      "grad_norm": 16.452594757080078,
      "learning_rate": 2.8598218471636194e-05,
      "loss": 4.0338,
      "step": 2300
    },
    {
      "epoch": 1.518506801645049,
      "grad_norm": 20.27753257751465,
      "learning_rate": 2.7426160337552742e-05,
      "loss": 4.0962,
      "step": 2400
    },
    {
      "epoch": 1.5817779183802594,
      "grad_norm": 15.492554664611816,
      "learning_rate": 2.6254102203469293e-05,
      "loss": 4.057,
      "step": 2500
    },
    {
      "epoch": 1.6450490351154698,
      "grad_norm": 15.298516273498535,
      "learning_rate": 2.508204406938584e-05,
      "loss": 4.0301,
      "step": 2600
    },
    {
      "epoch": 1.70832015185068,
      "grad_norm": 15.670785903930664,
      "learning_rate": 2.3909985935302392e-05,
      "loss": 4.0129,
      "step": 2700
    },
    {
      "epoch": 1.7715912685858906,
      "grad_norm": 18.541555404663086,
      "learning_rate": 2.2737927801218943e-05,
      "loss": 3.9724,
      "step": 2800
    },
    {
      "epoch": 1.834862385321101,
      "grad_norm": 19.13411521911621,
      "learning_rate": 2.156586966713549e-05,
      "loss": 4.0044,
      "step": 2900
    },
    {
      "epoch": 1.8981335020563113,
      "grad_norm": 14.532624244689941,
      "learning_rate": 2.039381153305204e-05,
      "loss": 3.9882,
      "step": 3000
    },
    {
      "epoch": 1.8981335020563113,
      "eval_runtime": 33.1181,
      "eval_samples_per_second": 95.416,
      "eval_steps_per_second": 11.927,
      "step": 3000
    },
    {
      "epoch": 1.9614046187915217,
      "grad_norm": 15.767202377319336,
      "learning_rate": 1.922175339896859e-05,
      "loss": 3.9372,
      "step": 3100
    },
    {
      "epoch": 2.024675735526732,
      "grad_norm": 17.210546493530273,
      "learning_rate": 1.804969526488514e-05,
      "loss": 3.9757,
      "step": 3200
    },
    {
      "epoch": 2.0879468522619424,
      "grad_norm": 15.209254264831543,
      "learning_rate": 1.6877637130801688e-05,
      "loss": 3.9668,
      "step": 3300
    },
    {
      "epoch": 2.1512179689971527,
      "grad_norm": 16.821176528930664,
      "learning_rate": 1.570557899671824e-05,
      "loss": 3.9732,
      "step": 3400
    },
    {
      "epoch": 2.2144890857323634,
      "grad_norm": 15.914960861206055,
      "learning_rate": 1.4533520862634786e-05,
      "loss": 3.9375,
      "step": 3500
    },
    {
      "epoch": 2.2777602024675736,
      "grad_norm": 16.489627838134766,
      "learning_rate": 1.3361462728551336e-05,
      "loss": 3.9993,
      "step": 3600
    },
    {
      "epoch": 2.341031319202784,
      "grad_norm": 17.943021774291992,
      "learning_rate": 1.2189404594467887e-05,
      "loss": 3.9889,
      "step": 3700
    },
    {
      "epoch": 2.4043024359379945,
      "grad_norm": 14.150239944458008,
      "learning_rate": 1.1017346460384436e-05,
      "loss": 4.0,
      "step": 3800
    },
    {
      "epoch": 2.4675735526732048,
      "grad_norm": 15.843707084655762,
      "learning_rate": 9.845288326300985e-06,
      "loss": 3.9527,
      "step": 3900
    },
    {
      "epoch": 2.530844669408415,
      "grad_norm": 16.922142028808594,
      "learning_rate": 8.673230192217533e-06,
      "loss": 3.8801,
      "step": 4000
    },
    {
      "epoch": 2.530844669408415,
      "eval_runtime": 33.0326,
      "eval_samples_per_second": 95.663,
      "eval_steps_per_second": 11.958,
      "step": 4000
    },
    {
      "epoch": 2.5941157861436253,
      "grad_norm": 19.046831130981445,
      "learning_rate": 7.501172058134085e-06,
      "loss": 3.9705,
      "step": 4100
    },
    {
      "epoch": 2.657386902878836,
      "grad_norm": 15.516547203063965,
      "learning_rate": 6.329113924050633e-06,
      "loss": 3.9296,
      "step": 4200
    },
    {
      "epoch": 2.720658019614046,
      "grad_norm": 16.0513916015625,
      "learning_rate": 5.157055789967183e-06,
      "loss": 4.0594,
      "step": 4300
    },
    {
      "epoch": 2.7839291363492564,
      "grad_norm": 13.820748329162598,
      "learning_rate": 3.984997655883732e-06,
      "loss": 3.9353,
      "step": 4400
    },
    {
      "epoch": 2.847200253084467,
      "grad_norm": 14.607758522033691,
      "learning_rate": 2.8129395218002813e-06,
      "loss": 4.0605,
      "step": 4500
    },
    {
      "epoch": 2.9104713698196774,
      "grad_norm": 18.928266525268555,
      "learning_rate": 1.6408813877168308e-06,
      "loss": 3.9663,
      "step": 4600
    },
    {
      "epoch": 2.9737424865548876,
      "grad_norm": 17.247835159301758,
      "learning_rate": 4.688232536333803e-07,
      "loss": 3.9057,
      "step": 4700
    }
  ],
  "logging_steps": 100,
  "max_steps": 4740,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2016724755342822.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}