{
  "best_global_step": 15000,
  "best_metric": 0.8307942152023315,
  "best_model_checkpoint": "ActionBert2-109M/checkpoint-15000",
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 16875,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 3.1670663356781006,
      "learning_rate": 1.8821925925925927e-05,
      "loss": 0.3813,
      "step": 1000
    },
    {
      "epoch": 0.17777777777777778,
      "eval_loss": 0.2506166696548462,
      "eval_mae": 0.3367820978164673,
      "eval_mse": 0.2506166696548462,
      "eval_r2": 0.7637249827384949,
      "eval_runtime": 4.791,
      "eval_samples_per_second": 2087.248,
      "eval_steps_per_second": 130.453,
      "step": 1000
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 3.0615577697753906,
      "learning_rate": 1.7636740740740744e-05,
      "loss": 0.263,
      "step": 2000
    },
    {
      "epoch": 0.35555555555555557,
      "eval_loss": 0.2408183068037033,
      "eval_mae": 0.31582972407341003,
      "eval_mse": 0.24081829190254211,
      "eval_r2": 0.7729626297950745,
      "eval_runtime": 4.7936,
      "eval_samples_per_second": 2086.113,
      "eval_steps_per_second": 130.382,
      "step": 2000
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 11.606523513793945,
      "learning_rate": 1.6451555555555554e-05,
      "loss": 0.2506,
      "step": 3000
    },
    {
      "epoch": 0.5333333333333333,
      "eval_loss": 0.21803471446037292,
      "eval_mae": 0.29080167412757874,
      "eval_mse": 0.21803471446037292,
      "eval_r2": 0.7944424152374268,
      "eval_runtime": 4.7889,
      "eval_samples_per_second": 2088.155,
      "eval_steps_per_second": 130.51,
      "step": 3000
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 13.706526756286621,
      "learning_rate": 1.5266370370370372e-05,
      "loss": 0.2414,
      "step": 4000
    },
    {
      "epoch": 0.7111111111111111,
      "eval_loss": 0.20693176984786987,
      "eval_mae": 0.2782182991504669,
      "eval_mse": 0.20693176984786987,
      "eval_r2": 0.8049100041389465,
      "eval_runtime": 4.7925,
      "eval_samples_per_second": 2086.596,
      "eval_steps_per_second": 130.412,
      "step": 4000
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 6.0101423263549805,
      "learning_rate": 1.4081185185185186e-05,
      "loss": 0.2295,
      "step": 5000
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 0.2030615508556366,
      "eval_mae": 0.27224966883659363,
      "eval_mse": 0.2030615508556366,
      "eval_r2": 0.8085587024688721,
      "eval_runtime": 4.7946,
      "eval_samples_per_second": 2085.665,
      "eval_steps_per_second": 130.354,
      "step": 5000
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 7.812684059143066,
      "learning_rate": 1.2896000000000002e-05,
      "loss": 0.1969,
      "step": 6000
    },
    {
      "epoch": 1.0666666666666667,
      "eval_loss": 0.19701267778873444,
      "eval_mae": 0.2610624134540558,
      "eval_mse": 0.19701267778873444,
      "eval_r2": 0.8142614364624023,
      "eval_runtime": 4.7965,
      "eval_samples_per_second": 2084.853,
      "eval_steps_per_second": 130.303,
      "step": 6000
    },
    {
      "epoch": 1.2444444444444445,
      "grad_norm": 4.992111682891846,
      "learning_rate": 1.1710814814814816e-05,
      "loss": 0.1495,
      "step": 7000
    },
    {
      "epoch": 1.2444444444444445,
      "eval_loss": 0.1891198754310608,
      "eval_mae": 0.24995234608650208,
      "eval_mse": 0.1891198605298996,
      "eval_r2": 0.8217025995254517,
      "eval_runtime": 4.7971,
      "eval_samples_per_second": 2084.602,
      "eval_steps_per_second": 130.288,
      "step": 7000
    },
    {
      "epoch": 1.4222222222222223,
      "grad_norm": 4.602079391479492,
      "learning_rate": 1.052562962962963e-05,
      "loss": 0.1557,
      "step": 8000
    },
    {
      "epoch": 1.4222222222222223,
      "eval_loss": 0.18423964083194733,
      "eval_mae": 0.2460847944021225,
      "eval_mse": 0.18423964083194733,
      "eval_r2": 0.8263035416603088,
      "eval_runtime": 4.7986,
      "eval_samples_per_second": 2083.948,
      "eval_steps_per_second": 130.247,
      "step": 8000
    },
    {
      "epoch": 1.6,
      "grad_norm": 4.8106303215026855,
      "learning_rate": 9.340444444444445e-06,
      "loss": 0.1474,
      "step": 9000
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.18420642614364624,
      "eval_mae": 0.24716965854167938,
      "eval_mse": 0.18420642614364624,
      "eval_r2": 0.8263348340988159,
      "eval_runtime": 4.7984,
      "eval_samples_per_second": 2084.015,
      "eval_steps_per_second": 130.251,
      "step": 9000
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 8.563740730285645,
      "learning_rate": 8.156444444444444e-06,
      "loss": 0.1419,
      "step": 10000
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 0.18527014553546906,
      "eval_mae": 0.24834123253822327,
      "eval_mse": 0.18527014553546906,
      "eval_r2": 0.8253320455551147,
      "eval_runtime": 4.7987,
      "eval_samples_per_second": 2083.902,
      "eval_steps_per_second": 130.244,
      "step": 10000
    },
    {
      "epoch": 1.9555555555555557,
      "grad_norm": 7.638633728027344,
      "learning_rate": 6.97125925925926e-06,
      "loss": 0.1513,
      "step": 11000
    },
    {
      "epoch": 1.9555555555555557,
      "eval_loss": 0.19055569171905518,
      "eval_mae": 0.2580980956554413,
      "eval_mse": 0.19055569171905518,
      "eval_r2": 0.8203489184379578,
      "eval_runtime": 4.7998,
      "eval_samples_per_second": 2083.401,
      "eval_steps_per_second": 130.213,
      "step": 11000
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 4.0671234130859375,
      "learning_rate": 5.78725925925926e-06,
      "loss": 0.1108,
      "step": 12000
    },
    {
      "epoch": 2.1333333333333333,
      "eval_loss": 0.1817326694726944,
      "eval_mae": 0.23058335483074188,
      "eval_mse": 0.1817326694726944,
      "eval_r2": 0.8286670446395874,
      "eval_runtime": 4.8008,
      "eval_samples_per_second": 2082.976,
      "eval_steps_per_second": 130.186,
      "step": 12000
    },
    {
      "epoch": 2.311111111111111,
      "grad_norm": 5.488609790802002,
      "learning_rate": 4.602074074074074e-06,
      "loss": 0.1026,
      "step": 13000
    },
    {
      "epoch": 2.311111111111111,
      "eval_loss": 0.18057070672512054,
      "eval_mae": 0.22670522332191467,
      "eval_mse": 0.18057070672512054,
      "eval_r2": 0.8297625184059143,
      "eval_runtime": 4.7977,
      "eval_samples_per_second": 2084.337,
      "eval_steps_per_second": 130.271,
      "step": 13000
    },
    {
      "epoch": 2.488888888888889,
      "grad_norm": 1.5936788320541382,
      "learning_rate": 3.416888888888889e-06,
      "loss": 0.0945,
      "step": 14000
    },
    {
      "epoch": 2.488888888888889,
      "eval_loss": 0.18265259265899658,
      "eval_mae": 0.2341051697731018,
      "eval_mse": 0.1826525777578354,
      "eval_r2": 0.8277997970581055,
      "eval_runtime": 4.8018,
      "eval_samples_per_second": 2082.558,
      "eval_steps_per_second": 130.16,
      "step": 14000
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 3.25566029548645,
      "learning_rate": 2.231703703703704e-06,
      "loss": 0.0936,
      "step": 15000
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 0.17947633564472198,
      "eval_mae": 0.22917917370796204,
      "eval_mse": 0.17947635054588318,
      "eval_r2": 0.8307942152023315,
      "eval_runtime": 4.7983,
      "eval_samples_per_second": 2084.067,
      "eval_steps_per_second": 130.254,
      "step": 15000
    },
    {
      "epoch": 2.8444444444444446,
      "grad_norm": 2.2979111671447754,
      "learning_rate": 1.0477037037037039e-06,
      "loss": 0.0978,
      "step": 16000
    },
    {
      "epoch": 2.8444444444444446,
      "eval_loss": 0.1797763854265213,
      "eval_mae": 0.2279701679944992,
      "eval_mse": 0.1797763854265213,
      "eval_r2": 0.8305113911628723,
      "eval_runtime": 4.8025,
      "eval_samples_per_second": 2082.257,
      "eval_steps_per_second": 130.141,
      "step": 16000
    },
    {
      "epoch": 3.0,
      "step": 16875,
      "total_flos": 3.551967355392e+16,
      "train_loss": 0.17141379213686342,
      "train_runtime": 610.1774,
      "train_samples_per_second": 442.494,
      "train_steps_per_second": 27.656
    }
  ],
  "logging_steps": 1000,
  "max_steps": 16875,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 3.551967355392e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}