{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 0.2216796875,
      "learning_rate": 4.943181818181818e-05,
      "loss": 0.5826,
      "step": 50
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 0.150390625,
      "learning_rate": 4.886363636363637e-05,
      "loss": 0.3128,
      "step": 100
    },
    {
      "epoch": 0.6818181818181818,
      "grad_norm": 0.1298828125,
      "learning_rate": 4.829545454545455e-05,
      "loss": 0.289,
      "step": 150
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.1435546875,
      "learning_rate": 4.772727272727273e-05,
      "loss": 0.2702,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.24845334887504578,
      "eval_runtime": 52.787,
      "eval_samples_per_second": 14.89,
      "eval_steps_per_second": 0.474,
      "step": 220
    },
    {
      "epoch": 1.1363636363636362,
      "grad_norm": 0.1396484375,
      "learning_rate": 4.715909090909091e-05,
      "loss": 0.2663,
      "step": 250
    },
    {
      "epoch": 1.3636363636363638,
      "grad_norm": 0.166015625,
      "learning_rate": 4.659090909090909e-05,
      "loss": 0.2591,
      "step": 300
    },
    {
      "epoch": 1.5909090909090908,
      "grad_norm": 0.162109375,
      "learning_rate": 4.602272727272727e-05,
      "loss": 0.2498,
      "step": 350
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.119140625,
      "learning_rate": 4.545454545454546e-05,
      "loss": 0.2557,
      "step": 400
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.2350398302078247,
      "eval_runtime": 53.0739,
      "eval_samples_per_second": 14.81,
      "eval_steps_per_second": 0.471,
      "step": 440
    },
    {
      "epoch": 2.0454545454545454,
      "grad_norm": 0.1259765625,
      "learning_rate": 4.488636363636364e-05,
      "loss": 0.2466,
      "step": 450
    },
    {
      "epoch": 2.2727272727272725,
      "grad_norm": 0.1416015625,
      "learning_rate": 4.431818181818182e-05,
      "loss": 0.2361,
      "step": 500
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.1337890625,
      "learning_rate": 4.375e-05,
      "loss": 0.253,
      "step": 550
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.1513671875,
      "learning_rate": 4.318181818181819e-05,
      "loss": 0.2409,
      "step": 600
    },
    {
      "epoch": 2.9545454545454546,
      "grad_norm": 0.1640625,
      "learning_rate": 4.261363636363637e-05,
      "loss": 0.2436,
      "step": 650
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.22992941737174988,
      "eval_runtime": 53.1353,
      "eval_samples_per_second": 14.792,
      "eval_steps_per_second": 0.47,
      "step": 660
    },
    {
      "epoch": 3.1818181818181817,
      "grad_norm": 0.16015625,
      "learning_rate": 4.204545454545455e-05,
      "loss": 0.2363,
      "step": 700
    },
    {
      "epoch": 3.409090909090909,
      "grad_norm": 0.1650390625,
      "learning_rate": 4.1477272727272734e-05,
      "loss": 0.2256,
      "step": 750
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.158203125,
      "learning_rate": 4.0909090909090915e-05,
      "loss": 0.2389,
      "step": 800
    },
    {
      "epoch": 3.8636363636363638,
      "grad_norm": 0.169921875,
      "learning_rate": 4.034090909090909e-05,
      "loss": 0.2291,
      "step": 850
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.22730520367622375,
      "eval_runtime": 53.1918,
      "eval_samples_per_second": 14.777,
      "eval_steps_per_second": 0.47,
      "step": 880
    },
    {
      "epoch": 4.090909090909091,
      "grad_norm": 0.1708984375,
      "learning_rate": 3.9772727272727275e-05,
      "loss": 0.2318,
      "step": 900
    },
    {
      "epoch": 4.318181818181818,
      "grad_norm": 0.181640625,
      "learning_rate": 3.9204545454545456e-05,
      "loss": 0.2322,
      "step": 950
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.1591796875,
      "learning_rate": 3.8636363636363636e-05,
      "loss": 0.2194,
      "step": 1000
    },
    {
      "epoch": 4.7727272727272725,
      "grad_norm": 0.166015625,
      "learning_rate": 3.8068181818181816e-05,
      "loss": 0.2277,
      "step": 1050
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.1767578125,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.2329,
      "step": 1100
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.2264343947172165,
      "eval_runtime": 54.0094,
      "eval_samples_per_second": 14.553,
      "eval_steps_per_second": 0.463,
      "step": 1100
    },
    {
      "epoch": 5.2272727272727275,
      "grad_norm": 0.1806640625,
      "learning_rate": 3.6931818181818184e-05,
      "loss": 0.2234,
      "step": 1150
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.19140625,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.2154,
      "step": 1200
    },
    {
      "epoch": 5.681818181818182,
      "grad_norm": 0.173828125,
      "learning_rate": 3.579545454545455e-05,
      "loss": 0.2349,
      "step": 1250
    },
    {
      "epoch": 5.909090909090909,
      "grad_norm": 0.162109375,
      "learning_rate": 3.522727272727273e-05,
      "loss": 0.2172,
      "step": 1300
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.22602269053459167,
      "eval_runtime": 54.0842,
      "eval_samples_per_second": 14.533,
      "eval_steps_per_second": 0.462,
      "step": 1320
    }
  ],
  "logging_steps": 50,
  "max_steps": 4400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 9.285391673846661e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}