{
  "best_global_step": 100,
  "best_metric": 0.581007719039917,
  "best_model_checkpoint": "outputs/lfm25-350m-commonsenseqa-lr1e4-r16-2k/checkpoint-100",
  "epoch": 2.0,
  "eval_steps": 50,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.49207068714313207,
      "epoch": 0.08,
      "grad_norm": 4.990988731384277,
      "learning_rate": 9.999578688879086e-05,
      "loss": 2.135391044616699,
      "mean_token_accuracy": 0.496875,
      "num_tokens": 9256.0,
      "step": 10
    },
    {
      "entropy": 0.7180297434329986,
      "epoch": 0.16,
      "grad_norm": 7.038885116577148,
      "learning_rate": 9.949107209404665e-05,
      "loss": 0.6821131229400634,
      "mean_token_accuracy": 0.7125,
      "num_tokens": 18706.0,
      "step": 20
    },
    {
      "entropy": 0.6099569849669934,
      "epoch": 0.24,
      "grad_norm": 9.42757511138916,
      "learning_rate": 9.815347070757232e-05,
      "loss": 0.7423738956451416,
      "mean_token_accuracy": 0.740625,
      "num_tokens": 27989.0,
      "step": 30
    },
    {
      "entropy": 0.6278413623571396,
      "epoch": 0.32,
      "grad_norm": 8.080451965332031,
      "learning_rate": 9.60054932593026e-05,
      "loss": 0.6707581043243408,
      "mean_token_accuracy": 0.73125,
      "num_tokens": 37273.0,
      "step": 40
    },
    {
      "entropy": 0.5547127045691014,
      "epoch": 0.4,
      "grad_norm": 5.448123931884766,
      "learning_rate": 9.30832881198487e-05,
      "loss": 0.6080288887023926,
      "mean_token_accuracy": 0.759375,
      "num_tokens": 46599.0,
      "step": 50
    },
    {
      "epoch": 0.4,
      "eval_entropy": 0.6053476703712364,
      "eval_loss": 0.6080912351608276,
      "eval_mean_token_accuracy": 0.7634803921568627,
      "eval_num_tokens": 46599.0,
      "eval_runtime": 11.5175,
      "eval_samples_per_second": 106.013,
      "eval_steps_per_second": 26.568,
      "step": 50
    },
    {
      "entropy": 0.6223195999860763,
      "epoch": 0.48,
      "grad_norm": 5.2335968017578125,
      "learning_rate": 8.943603315859101e-05,
      "loss": 0.6296503067016601,
      "mean_token_accuracy": 0.775,
      "num_tokens": 55998.0,
      "step": 60
    },
    {
      "entropy": 0.5325490891933441,
      "epoch": 0.56,
      "grad_norm": 3.735042095184326,
      "learning_rate": 8.512510812798426e-05,
      "loss": 0.6199626445770263,
      "mean_token_accuracy": 0.771875,
      "num_tokens": 65297.0,
      "step": 70
    },
    {
      "entropy": 0.6412477873265743,
      "epoch": 0.64,
      "grad_norm": 4.131916046142578,
      "learning_rate": 8.022306170204233e-05,
      "loss": 0.6183301448822022,
      "mean_token_accuracy": 0.759375,
      "num_tokens": 74530.0,
      "step": 80
    },
    {
      "entropy": 0.5066440511494875,
      "epoch": 0.72,
      "grad_norm": 6.953462600708008,
      "learning_rate": 7.481239055273959e-05,
      "loss": 0.6243555068969726,
      "mean_token_accuracy": 0.746875,
      "num_tokens": 83810.0,
      "step": 90
    },
    {
      "entropy": 0.5975224792957305,
      "epoch": 0.8,
      "grad_norm": 5.960926055908203,
      "learning_rate": 6.89841510112857e-05,
      "loss": 0.6379575729370117,
      "mean_token_accuracy": 0.75,
      "num_tokens": 93098.0,
      "step": 100
    },
    {
      "epoch": 0.8,
      "eval_entropy": 0.6016070044118594,
      "eval_loss": 0.581007719039917,
      "eval_mean_token_accuracy": 0.7745098039215687,
      "eval_num_tokens": 93098.0,
      "eval_runtime": 11.377,
      "eval_samples_per_second": 107.322,
      "eval_steps_per_second": 26.896,
      "step": 100
    },
    {
      "entropy": 0.593381131440401,
      "epoch": 0.88,
      "grad_norm": 5.572819232940674,
      "learning_rate": 6.283642667866316e-05,
      "loss": 0.5591686725616455,
      "mean_token_accuracy": 0.803125,
      "num_tokens": 102498.0,
      "step": 110
    },
    {
      "entropy": 0.5462032213807106,
      "epoch": 0.96,
      "grad_norm": 5.610764980316162,
      "learning_rate": 5.647267777405177e-05,
      "loss": 0.5735011577606202,
      "mean_token_accuracy": 0.784375,
      "num_tokens": 111914.0,
      "step": 120
    },
    {
      "entropy": 0.5103660989552736,
      "epoch": 1.04,
      "grad_norm": 4.618079662322998,
      "learning_rate": 5e-05,
      "loss": 0.4889837265014648,
      "mean_token_accuracy": 0.821875,
      "num_tokens": 121287.0,
      "step": 130
    },
    {
      "entropy": 0.2951751688495278,
      "epoch": 1.12,
      "grad_norm": 11.136743545532227,
      "learning_rate": 4.3527322225948234e-05,
      "loss": 0.3513810634613037,
      "mean_token_accuracy": 0.89375,
      "num_tokens": 130667.0,
      "step": 140
    },
    {
      "entropy": 0.2440035931766033,
      "epoch": 1.2,
      "grad_norm": 8.998021125793457,
      "learning_rate": 3.716357332133687e-05,
      "loss": 0.3495033740997314,
      "mean_token_accuracy": 0.9,
      "num_tokens": 139891.0,
      "step": 150
    },
    {
      "epoch": 1.2,
      "eval_entropy": 0.3550063371658325,
      "eval_loss": 0.6342237591743469,
      "eval_mean_token_accuracy": 0.7879901960784313,
      "eval_num_tokens": 139891.0,
      "eval_runtime": 11.4365,
      "eval_samples_per_second": 106.763,
      "eval_steps_per_second": 26.756,
      "step": 150
    },
    {
      "entropy": 0.33935176096856595,
      "epoch": 1.28,
      "grad_norm": 4.766120433807373,
      "learning_rate": 3.1015848988714305e-05,
      "loss": 0.3303521156311035,
      "mean_token_accuracy": 0.8875,
      "num_tokens": 149299.0,
      "step": 160
    },
    {
      "entropy": 0.36539566218853,
      "epoch": 1.3599999999999999,
      "grad_norm": 7.436731338500977,
      "learning_rate": 2.5187609447260417e-05,
      "loss": 0.348486065864563,
      "mean_token_accuracy": 0.865625,
      "num_tokens": 158544.0,
      "step": 170
    },
    {
      "entropy": 0.33908074218779805,
      "epoch": 1.44,
      "grad_norm": 8.012598991394043,
      "learning_rate": 1.9776938297957687e-05,
      "loss": 0.4287095069885254,
      "mean_token_accuracy": 0.834375,
      "num_tokens": 167960.0,
      "step": 180
    },
    {
      "entropy": 0.3853272207081318,
      "epoch": 1.52,
      "grad_norm": 5.60869836807251,
      "learning_rate": 1.4874891872015734e-05,
      "loss": 0.3443429708480835,
      "mean_token_accuracy": 0.890625,
      "num_tokens": 177219.0,
      "step": 190
    },
    {
      "entropy": 0.3327586879953742,
      "epoch": 1.6,
      "grad_norm": 6.2800164222717285,
      "learning_rate": 1.0563966841408995e-05,
      "loss": 0.3429755926132202,
      "mean_token_accuracy": 0.89375,
      "num_tokens": 186610.0,
      "step": 200
    },
    {
      "epoch": 1.6,
      "eval_entropy": 0.37273075129057653,
      "eval_loss": 0.6045761704444885,
      "eval_mean_token_accuracy": 0.7920751633986928,
      "eval_num_tokens": 186610.0,
      "eval_runtime": 11.4402,
      "eval_samples_per_second": 106.729,
      "eval_steps_per_second": 26.748,
      "step": 200
    },
    {
      "entropy": 0.3120275104418397,
      "epoch": 1.6800000000000002,
      "grad_norm": 3.8584883213043213,
      "learning_rate": 6.916711880151306e-06,
      "loss": 0.35584330558776855,
      "mean_token_accuracy": 0.871875,
      "num_tokens": 196043.0,
      "step": 210
    },
    {
      "entropy": 0.32570266518741847,
      "epoch": 1.76,
      "grad_norm": 7.246922492980957,
      "learning_rate": 3.994506740697407e-06,
      "loss": 0.3419358730316162,
      "mean_token_accuracy": 0.878125,
      "num_tokens": 205352.0,
      "step": 220
    },
    {
      "entropy": 0.31593060959130526,
      "epoch": 1.8399999999999999,
      "grad_norm": 5.6120829582214355,
      "learning_rate": 1.8465292924276845e-06,
      "loss": 0.3507185935974121,
      "mean_token_accuracy": 0.88125,
      "num_tokens": 214621.0,
      "step": 230
    },
    {
      "entropy": 0.34746452011168005,
      "epoch": 1.92,
      "grad_norm": 6.788421154022217,
      "learning_rate": 5.089279059533658e-07,
      "loss": 0.37170000076293946,
      "mean_token_accuracy": 0.86875,
      "num_tokens": 223936.0,
      "step": 240
    },
    {
      "entropy": 0.3478228982537985,
      "epoch": 2.0,
      "grad_norm": 4.587532997131348,
      "learning_rate": 4.213111209155907e-09,
      "loss": 0.29353837966918944,
      "mean_token_accuracy": 0.8875,
      "num_tokens": 233188.0,
      "step": 250
    },
    {
      "epoch": 2.0,
      "eval_entropy": 0.38705892378793044,
      "eval_loss": 0.5934809446334839,
      "eval_mean_token_accuracy": 0.7900326797385621,
      "eval_num_tokens": 233188.0,
      "eval_runtime": 11.4469,
      "eval_samples_per_second": 106.667,
      "eval_steps_per_second": 26.732,
      "step": 250
    },
    {
      "epoch": 2.0,
      "step": 250,
      "total_flos": 463827616407552.0,
      "train_loss": 0.5520024652481079,
      "train_runtime": 156.6255,
      "train_samples_per_second": 25.539,
      "train_steps_per_second": 1.596
    }
  ],
  "logging_steps": 10,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 463827616407552.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}