{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 50,
  "global_step": 1701,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11757789535567313,
      "grad_norm": 1.5434420108795166,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 0.7456,
      "step": 50
    },
    {
      "epoch": 0.23515579071134626,
      "grad_norm": 0.963267982006073,
      "learning_rate": 5.823529411764706e-06,
      "loss": 0.5784,
      "step": 100
    },
    {
      "epoch": 0.3527336860670194,
      "grad_norm": 1.1193718910217285,
      "learning_rate": 8.705882352941177e-06,
      "loss": 0.3618,
      "step": 150
    },
    {
      "epoch": 0.4703115814226925,
      "grad_norm": 0.7008563876152039,
      "learning_rate": 1.1647058823529412e-05,
      "loss": 0.3389,
      "step": 200
    },
    {
      "epoch": 0.5878894767783657,
      "grad_norm": 0.6045801639556885,
      "learning_rate": 1.4588235294117647e-05,
      "loss": 0.3334,
      "step": 250
    },
    {
      "epoch": 0.7054673721340388,
      "grad_norm": 0.5706155300140381,
      "learning_rate": 1.7529411764705884e-05,
      "loss": 0.2923,
      "step": 300
    },
    {
      "epoch": 0.823045267489712,
      "grad_norm": 0.6757872104644775,
      "learning_rate": 2.047058823529412e-05,
      "loss": 0.2901,
      "step": 350
    },
    {
      "epoch": 0.940623162845385,
      "grad_norm": 0.5101656317710876,
      "learning_rate": 2.3411764705882354e-05,
      "loss": 0.2893,
      "step": 400
    },
    {
      "epoch": 0.9994121105232217,
      "eval_loss": 0.4672001004219055,
      "eval_runtime": 174.9103,
      "eval_samples_per_second": 1.081,
      "eval_steps_per_second": 0.137,
      "step": 425
    },
    {
      "epoch": 1.0582010582010581,
      "grad_norm": 0.5830298066139221,
      "learning_rate": 2.6352941176470592e-05,
      "loss": 0.2832,
      "step": 450
    },
    {
      "epoch": 1.1757789535567313,
      "grad_norm": 0.5612862706184387,
      "learning_rate": 2.9294117647058827e-05,
      "loss": 0.2486,
      "step": 500
    },
    {
      "epoch": 1.2933568489124045,
      "grad_norm": 0.5751911997795105,
      "learning_rate": 3.223529411764706e-05,
      "loss": 0.2924,
      "step": 550
    },
    {
      "epoch": 1.4109347442680775,
      "grad_norm": 0.6772521734237671,
      "learning_rate": 3.5176470588235294e-05,
      "loss": 0.2899,
      "step": 600
    },
    {
      "epoch": 1.5285126396237507,
      "grad_norm": 0.6256781220436096,
      "learning_rate": 3.811764705882353e-05,
      "loss": 0.2733,
      "step": 650
    },
    {
      "epoch": 1.646090534979424,
      "grad_norm": 0.7451750040054321,
      "learning_rate": 4.1058823529411764e-05,
      "loss": 0.2874,
      "step": 700
    },
    {
      "epoch": 1.763668430335097,
      "grad_norm": 0.6674547791481018,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2668,
      "step": 750
    },
    {
      "epoch": 1.88124632569077,
      "grad_norm": 0.5857487916946411,
      "learning_rate": 4.694117647058824e-05,
      "loss": 0.2699,
      "step": 800
    },
    {
      "epoch": 1.9988242210464433,
      "grad_norm": 0.4972004294395447,
      "learning_rate": 4.9882352941176476e-05,
      "loss": 0.2867,
      "step": 850
    },
    {
      "epoch": 1.9988242210464433,
      "eval_loss": 0.43864762783050537,
      "eval_runtime": 172.7658,
      "eval_samples_per_second": 1.094,
      "eval_steps_per_second": 0.139,
      "step": 850
    },
    {
      "epoch": 2.1164021164021163,
      "grad_norm": 0.43465057015419006,
      "learning_rate": 4.929411764705882e-05,
      "loss": 0.2874,
      "step": 900
    },
    {
      "epoch": 2.2339800117577897,
      "grad_norm": 0.45625609159469604,
      "learning_rate": 4.855882352941177e-05,
      "loss": 0.2549,
      "step": 950
    },
    {
      "epoch": 2.3515579071134627,
      "grad_norm": 0.4708680510520935,
      "learning_rate": 4.7823529411764704e-05,
      "loss": 0.2435,
      "step": 1000
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 0.4648517370223999,
      "learning_rate": 4.708823529411765e-05,
      "loss": 0.2397,
      "step": 1050
    },
    {
      "epoch": 2.586713697824809,
      "grad_norm": 0.39246800541877747,
      "learning_rate": 4.635294117647059e-05,
      "loss": 0.2676,
      "step": 1100
    },
    {
      "epoch": 2.704291593180482,
      "grad_norm": 0.579401433467865,
      "learning_rate": 4.5617647058823535e-05,
      "loss": 0.2207,
      "step": 1150
    },
    {
      "epoch": 2.821869488536155,
      "grad_norm": 0.3959507942199707,
      "learning_rate": 4.4882352941176476e-05,
      "loss": 0.2247,
      "step": 1200
    },
    {
      "epoch": 2.9394473838918285,
      "grad_norm": 0.3841804265975952,
      "learning_rate": 4.414705882352941e-05,
      "loss": 0.2639,
      "step": 1250
    },
    {
      "epoch": 2.998236331569665,
      "eval_loss": 0.42491579055786133,
      "eval_runtime": 172.7491,
      "eval_samples_per_second": 1.094,
      "eval_steps_per_second": 0.139,
      "step": 1275
    },
    {
      "epoch": 3.0570252792475014,
      "grad_norm": 0.3887556791305542,
      "learning_rate": 4.341176470588236e-05,
      "loss": 0.2318,
      "step": 1300
    },
    {
      "epoch": 3.1746031746031744,
      "grad_norm": 0.2942847013473511,
      "learning_rate": 4.267647058823529e-05,
      "loss": 0.2351,
      "step": 1350
    },
    {
      "epoch": 3.292181069958848,
      "grad_norm": 0.41963356733322144,
      "learning_rate": 4.194117647058824e-05,
      "loss": 0.2449,
      "step": 1400
    },
    {
      "epoch": 3.409758965314521,
      "grad_norm": 0.41531622409820557,
      "learning_rate": 4.1205882352941176e-05,
      "loss": 0.2484,
      "step": 1450
    },
    {
      "epoch": 3.527336860670194,
      "grad_norm": 0.35044431686401367,
      "learning_rate": 4.0470588235294124e-05,
      "loss": 0.2331,
      "step": 1500
    },
    {
      "epoch": 3.6449147560258672,
      "grad_norm": 0.3828336000442505,
      "learning_rate": 3.973529411764706e-05,
      "loss": 0.2287,
      "step": 1550
    },
    {
      "epoch": 3.76249265138154,
      "grad_norm": 0.4572592079639435,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.2297,
      "step": 1600
    },
    {
      "epoch": 3.8800705467372136,
      "grad_norm": 0.47478222846984863,
      "learning_rate": 3.826470588235294e-05,
      "loss": 0.2285,
      "step": 1650
    },
    {
      "epoch": 3.9976484420928866,
      "grad_norm": 0.39389339089393616,
      "learning_rate": 3.752941176470588e-05,
      "loss": 0.2096,
      "step": 1700
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.4230487048625946,
      "eval_runtime": 172.772,
      "eval_samples_per_second": 1.094,
      "eval_steps_per_second": 0.139,
      "step": 1701
    }
  ],
  "logging_steps": 50,
  "max_steps": 4250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 8.215788830190797e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}