{
  "best_global_step": 150,
  "best_metric": 1.1094907522201538,
  "best_model_checkpoint": "mistral-7b-frlegal-lora/checkpoint-100",
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 255,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11834319526627218,
      "grad_norm": 3.7848479747772217,
      "learning_rate": 6.153846153846155e-05,
      "loss": 2.8047,
      "step": 10
    },
    {
      "epoch": 0.23668639053254437,
      "grad_norm": 3.650196075439453,
      "learning_rate": 0.00013846153846153847,
      "loss": 1.8139,
      "step": 20
    },
    {
      "epoch": 0.35502958579881655,
      "grad_norm": 4.603827476501465,
      "learning_rate": 0.00019912663755458515,
      "loss": 1.5624,
      "step": 30
    },
    {
      "epoch": 0.47337278106508873,
      "grad_norm": 3.440032482147217,
      "learning_rate": 0.0001903930131004367,
      "loss": 1.3397,
      "step": 40
    },
    {
      "epoch": 0.591715976331361,
      "grad_norm": 4.5781025886535645,
      "learning_rate": 0.00018165938864628822,
      "loss": 1.2809,
      "step": 50
    },
    {
      "epoch": 0.591715976331361,
      "eval_loss": 1.2312296628952026,
      "eval_runtime": 60.6234,
      "eval_samples_per_second": 2.788,
      "eval_steps_per_second": 1.402,
      "step": 50
    },
    {
      "epoch": 0.7100591715976331,
      "grad_norm": 3.0015316009521484,
      "learning_rate": 0.00017292576419213976,
      "loss": 1.1795,
      "step": 60
    },
    {
      "epoch": 0.8284023668639053,
      "grad_norm": 2.660057544708252,
      "learning_rate": 0.00016419213973799127,
      "loss": 1.0804,
      "step": 70
    },
    {
      "epoch": 0.9467455621301775,
      "grad_norm": 2.6601247787475586,
      "learning_rate": 0.0001554585152838428,
      "loss": 1.1704,
      "step": 80
    },
    {
      "epoch": 1.0591715976331362,
      "grad_norm": 2.8257343769073486,
      "learning_rate": 0.00014672489082969434,
      "loss": 0.9553,
      "step": 90
    },
    {
      "epoch": 1.1775147928994083,
      "grad_norm": 2.2877657413482666,
      "learning_rate": 0.00013799126637554588,
      "loss": 0.8022,
      "step": 100
    },
    {
      "epoch": 1.1775147928994083,
      "eval_loss": 1.1341192722320557,
      "eval_runtime": 60.7323,
      "eval_samples_per_second": 2.783,
      "eval_steps_per_second": 1.4,
      "step": 100
    },
    {
      "epoch": 1.2958579881656804,
      "grad_norm": 2.4518744945526123,
      "learning_rate": 0.00012925764192139738,
      "loss": 0.7463,
      "step": 110
    },
    {
      "epoch": 1.4142011834319526,
      "grad_norm": 2.634495735168457,
      "learning_rate": 0.0001205240174672489,
      "loss": 0.7574,
      "step": 120
    },
    {
      "epoch": 1.532544378698225,
      "grad_norm": 2.428051710128784,
      "learning_rate": 0.00011179039301310044,
      "loss": 0.7658,
      "step": 130
    },
    {
      "epoch": 1.650887573964497,
      "grad_norm": 2.903827667236328,
      "learning_rate": 0.00010305676855895196,
      "loss": 0.7334,
      "step": 140
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 2.4579429626464844,
      "learning_rate": 9.43231441048035e-05,
      "loss": 0.7041,
      "step": 150
    },
    {
      "epoch": 1.7692307692307692,
      "eval_loss": 1.1094907522201538,
      "eval_runtime": 60.6543,
      "eval_samples_per_second": 2.786,
      "eval_steps_per_second": 1.401,
      "step": 150
    },
    {
      "epoch": 1.8875739644970415,
      "grad_norm": 2.4557547569274902,
      "learning_rate": 8.558951965065502e-05,
      "loss": 0.7263,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.915830135345459,
      "learning_rate": 7.685589519650656e-05,
      "loss": 0.784,
      "step": 170
    },
    {
      "epoch": 2.1183431952662723,
      "grad_norm": 3.7625837326049805,
      "learning_rate": 6.812227074235808e-05,
      "loss": 0.4613,
      "step": 180
    },
    {
      "epoch": 2.2366863905325443,
      "grad_norm": 2.049013376235962,
      "learning_rate": 5.9388646288209616e-05,
      "loss": 0.4533,
      "step": 190
    },
    {
      "epoch": 2.3550295857988166,
      "grad_norm": 1.8300621509552002,
      "learning_rate": 5.065502183406113e-05,
      "loss": 0.456,
      "step": 200
    },
    {
      "epoch": 2.3550295857988166,
      "eval_loss": 1.1557873487472534,
      "eval_runtime": 60.6355,
      "eval_samples_per_second": 2.787,
      "eval_steps_per_second": 1.402,
      "step": 200
    },
    {
      "epoch": 2.4733727810650885,
      "grad_norm": 1.6030356884002686,
      "learning_rate": 4.192139737991266e-05,
      "loss": 0.4833,
      "step": 210
    },
    {
      "epoch": 2.591715976331361,
      "grad_norm": 2.6860575675964355,
      "learning_rate": 3.3187772925764197e-05,
      "loss": 0.5044,
      "step": 220
    },
    {
      "epoch": 2.710059171597633,
      "grad_norm": 2.007375955581665,
      "learning_rate": 2.445414847161572e-05,
      "loss": 0.4456,
      "step": 230
    },
    {
      "epoch": 2.828402366863905,
      "grad_norm": 1.3755784034729004,
      "learning_rate": 1.572052401746725e-05,
      "loss": 0.4325,
      "step": 240
    },
    {
      "epoch": 2.9467455621301775,
      "grad_norm": 2.183783769607544,
      "learning_rate": 6.986899563318778e-06,
      "loss": 0.4677,
      "step": 250
    },
    {
      "epoch": 2.9467455621301775,
      "eval_loss": 1.1586920022964478,
      "eval_runtime": 60.6003,
      "eval_samples_per_second": 2.789,
      "eval_steps_per_second": 1.403,
      "step": 250
    }
  ],
  "logging_steps": 10,
  "max_steps": 255,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8132308222967808.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}