{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0476190476190474,
  "eval_steps": 3,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09523809523809523,
      "eval_loss": 1.5588610172271729,
      "eval_runtime": 1.0889,
      "eval_samples_per_second": 16.53,
      "eval_steps_per_second": 2.755,
      "step": 1
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.8132609128952026,
      "learning_rate": 3e-05,
      "loss": 1.5202,
      "step": 3
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 1.5518724918365479,
      "eval_runtime": 0.5151,
      "eval_samples_per_second": 34.943,
      "eval_steps_per_second": 5.824,
      "step": 3
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.6047109961509705,
      "learning_rate": 6e-05,
      "loss": 1.4653,
      "step": 6
    },
    {
      "epoch": 0.5714285714285714,
      "eval_loss": 1.4867305755615234,
      "eval_runtime": 0.5144,
      "eval_samples_per_second": 34.991,
      "eval_steps_per_second": 5.832,
      "step": 6
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.7678360342979431,
      "learning_rate": 9e-05,
      "loss": 1.435,
      "step": 9
    },
    {
      "epoch": 0.8571428571428571,
      "eval_loss": 1.4172625541687012,
      "eval_runtime": 0.5187,
      "eval_samples_per_second": 34.702,
      "eval_steps_per_second": 5.784,
      "step": 9
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.7624624967575073,
      "learning_rate": 9.797464868072488e-05,
      "loss": 1.5631,
      "step": 12
    },
    {
      "epoch": 1.1428571428571428,
      "eval_loss": 1.40391206741333,
      "eval_runtime": 0.5161,
      "eval_samples_per_second": 34.88,
      "eval_steps_per_second": 5.813,
      "step": 12
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.6058403849601746,
      "learning_rate": 8.778747871771292e-05,
      "loss": 1.1919,
      "step": 15
    },
    {
      "epoch": 1.4285714285714286,
      "eval_loss": 1.3703105449676514,
      "eval_runtime": 0.5153,
      "eval_samples_per_second": 34.93,
      "eval_steps_per_second": 5.822,
      "step": 15
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.6363664269447327,
      "learning_rate": 7.077075065009433e-05,
      "loss": 1.3437,
      "step": 18
    },
    {
      "epoch": 1.7142857142857144,
      "eval_loss": 1.3599408864974976,
      "eval_runtime": 0.5155,
      "eval_samples_per_second": 34.92,
      "eval_steps_per_second": 5.82,
      "step": 18
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0324137210845947,
      "learning_rate": 5e-05,
      "loss": 1.3446,
      "step": 21
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3587403297424316,
      "eval_runtime": 0.5143,
      "eval_samples_per_second": 35.0,
      "eval_steps_per_second": 5.833,
      "step": 21
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.6064738035202026,
      "learning_rate": 2.9229249349905684e-05,
      "loss": 1.0781,
      "step": 24
    },
    {
      "epoch": 2.2857142857142856,
      "eval_loss": 1.3659809827804565,
      "eval_runtime": 0.5145,
      "eval_samples_per_second": 34.989,
      "eval_steps_per_second": 5.831,
      "step": 24
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.5662018060684204,
      "learning_rate": 1.2212521282287092e-05,
      "loss": 1.0401,
      "step": 27
    },
    {
      "epoch": 2.571428571428571,
      "eval_loss": 1.3706698417663574,
      "eval_runtime": 0.5165,
      "eval_samples_per_second": 34.852,
      "eval_steps_per_second": 5.809,
      "step": 27
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.620604395866394,
      "learning_rate": 2.0253513192751373e-06,
      "loss": 1.0945,
      "step": 30
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 1.372444987297058,
      "eval_runtime": 0.5147,
      "eval_samples_per_second": 34.974,
      "eval_steps_per_second": 5.829,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8810685655941120.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}