{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.6923076923076925,
  "eval_steps": 5,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 2.1649128036114615,
      "learning_rate": 4e-05,
      "loss": 3.1505,
      "step": 1
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.4215155841281863,
      "learning_rate": 0.0002,
      "loss": 3.0231,
      "step": 5
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 2.6061668395996094,
      "eval_runtime": 2.6215,
      "eval_samples_per_second": 7.629,
      "eval_steps_per_second": 1.144,
      "step": 5
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.651519916913547,
      "learning_rate": 0.00019396926207859084,
      "loss": 1.8692,
      "step": 10
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 1.6896581649780273,
      "eval_runtime": 1.3449,
      "eval_samples_per_second": 14.871,
      "eval_steps_per_second": 2.231,
      "step": 10
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 1.9515671701626458,
      "learning_rate": 0.0001766044443118978,
      "loss": 1.3807,
      "step": 15
    },
    {
      "epoch": 2.3076923076923075,
      "eval_loss": 1.45188570022583,
      "eval_runtime": 1.3413,
      "eval_samples_per_second": 14.911,
      "eval_steps_per_second": 2.237,
      "step": 15
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.6039953656556071,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.165,
      "step": 20
    },
    {
      "epoch": 3.076923076923077,
      "eval_loss": 1.377772569656372,
      "eval_runtime": 1.3458,
      "eval_samples_per_second": 14.861,
      "eval_steps_per_second": 2.229,
      "step": 20
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.2836516448253057,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.0943,
      "step": 25
    },
    {
      "epoch": 3.8461538461538463,
      "eval_loss": 1.3182907104492188,
      "eval_runtime": 1.3427,
      "eval_samples_per_second": 14.895,
      "eval_steps_per_second": 2.234,
      "step": 25
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 0.30847224546147556,
      "learning_rate": 8.263518223330697e-05,
      "loss": 1.0287,
      "step": 30
    },
    {
      "epoch": 4.615384615384615,
      "eval_loss": 1.3018699884414673,
      "eval_runtime": 1.3484,
      "eval_samples_per_second": 14.832,
      "eval_steps_per_second": 2.225,
      "step": 30
    },
    {
      "epoch": 5.384615384615385,
      "grad_norm": 0.44903587891321267,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.9179,
      "step": 35
    },
    {
      "epoch": 5.384615384615385,
      "eval_loss": 1.3123514652252197,
      "eval_runtime": 1.3412,
      "eval_samples_per_second": 14.912,
      "eval_steps_per_second": 2.237,
      "step": 35
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 0.32136925605744115,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.9144,
      "step": 40
    },
    {
      "epoch": 6.153846153846154,
      "eval_loss": 1.325203776359558,
      "eval_runtime": 1.3435,
      "eval_samples_per_second": 14.886,
      "eval_steps_per_second": 2.233,
      "step": 40
    },
    {
      "epoch": 6.923076923076923,
      "grad_norm": 0.4351422163226237,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.879,
      "step": 45
    },
    {
      "epoch": 6.923076923076923,
      "eval_loss": 1.3339039087295532,
      "eval_runtime": 1.3424,
      "eval_samples_per_second": 14.898,
      "eval_steps_per_second": 2.235,
      "step": 45
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 0.3920332363997211,
      "learning_rate": 0.0,
      "loss": 0.8385,
      "step": 50
    },
    {
      "epoch": 7.6923076923076925,
      "eval_loss": 1.3345052003860474,
      "eval_runtime": 1.3405,
      "eval_samples_per_second": 14.92,
      "eval_steps_per_second": 2.238,
      "step": 50
    },
    {
      "epoch": 7.6923076923076925,
      "step": 50,
      "total_flos": 421095309312.0,
      "train_loss": 1.3136224746704102,
      "train_runtime": 178.8914,
      "train_samples_per_second": 4.472,
      "train_steps_per_second": 0.279
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 421095309312.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}