checkpoint-2400/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.1447721179624666,
  "eval_steps": 200,
  "global_step": 2400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17873100983020554,
      "grad_norm": 0.5410762429237366,
      "learning_rate": 9.999142157322298e-05,
      "loss": 0.5394,
      "step": 200
    },
    {
      "epoch": 0.17873100983020554,
      "eval_loss": 0.2514195740222931,
      "eval_runtime": 56.7459,
      "eval_samples_per_second": 3.19,
      "eval_steps_per_second": 1.604,
      "step": 200
    },
    {
      "epoch": 0.3574620196604111,
      "grad_norm": 0.3244825303554535,
      "learning_rate": 9.95497608932715e-05,
      "loss": 0.2114,
      "step": 400
    },
    {
      "epoch": 0.3574620196604111,
      "eval_loss": 0.19941778481006622,
      "eval_runtime": 56.7365,
      "eval_samples_per_second": 3.19,
      "eval_steps_per_second": 1.604,
      "step": 400
    },
    {
      "epoch": 0.5361930294906166,
      "grad_norm": 0.23322512209415436,
      "learning_rate": 9.844466810070319e-05,
      "loss": 0.184,
      "step": 600
    },
    {
      "epoch": 0.5361930294906166,
      "eval_loss": 0.1872645616531372,
      "eval_runtime": 56.8389,
      "eval_samples_per_second": 3.184,
      "eval_steps_per_second": 1.601,
      "step": 600
    },
    {
      "epoch": 0.7149240393208222,
      "grad_norm": 0.20280934870243073,
      "learning_rate": 9.669093951406317e-05,
      "loss": 0.1797,
      "step": 800
    },
    {
      "epoch": 0.7149240393208222,
      "eval_loss": 0.18097320199012756,
      "eval_runtime": 56.861,
      "eval_samples_per_second": 3.183,
      "eval_steps_per_second": 1.6,
      "step": 800
    },
    {
      "epoch": 0.8936550491510277,
      "grad_norm": 0.18255303800106049,
      "learning_rate": 9.431205617221452e-05,
      "loss": 0.1719,
      "step": 1000
    },
    {
      "epoch": 0.8936550491510277,
      "eval_loss": 0.17648382484912872,
      "eval_runtime": 58.6626,
      "eval_samples_per_second": 3.085,
      "eval_steps_per_second": 1.551,
      "step": 1000
    },
    {
      "epoch": 1.0723860589812333,
      "grad_norm": 0.15965856611728668,
      "learning_rate": 9.133986944179754e-05,
      "loss": 0.1647,
      "step": 1200
    },
    {
      "epoch": 1.0723860589812333,
      "eval_loss": 0.17539893090724945,
      "eval_runtime": 58.6505,
      "eval_samples_per_second": 3.086,
      "eval_steps_per_second": 1.552,
      "step": 1200
    },
    {
      "epoch": 1.2511170688114388,
      "grad_norm": 0.187672421336174,
      "learning_rate": 8.781417455261971e-05,
      "loss": 0.1538,
      "step": 1400
    },
    {
      "epoch": 1.2511170688114388,
      "eval_loss": 0.1730012148618698,
      "eval_runtime": 58.3686,
      "eval_samples_per_second": 3.101,
      "eval_steps_per_second": 1.559,
      "step": 1400
    },
    {
      "epoch": 1.4298480786416443,
      "grad_norm": 0.21108384430408478,
      "learning_rate": 8.378217777099999e-05,
      "loss": 0.1523,
      "step": 1600
    },
    {
      "epoch": 1.4298480786416443,
      "eval_loss": 0.17163121700286865,
      "eval_runtime": 58.6595,
      "eval_samples_per_second": 3.086,
      "eval_steps_per_second": 1.551,
      "step": 1600
    },
    {
      "epoch": 1.6085790884718498,
      "grad_norm": 0.1646961271762848,
      "learning_rate": 7.929786434519532e-05,
      "loss": 0.1524,
      "step": 1800
    },
    {
      "epoch": 1.6085790884718498,
      "eval_loss": 0.1694817841053009,
      "eval_runtime": 56.1831,
      "eval_samples_per_second": 3.222,
      "eval_steps_per_second": 1.62,
      "step": 1800
    },
    {
      "epoch": 1.7873100983020556,
      "grad_norm": 0.19810634851455688,
      "learning_rate": 7.442127568561985e-05,
      "loss": 0.1487,
      "step": 2000
    },
    {
      "epoch": 1.7873100983020556,
      "eval_loss": 0.1678919941186905,
      "eval_runtime": 56.3246,
      "eval_samples_per_second": 3.214,
      "eval_steps_per_second": 1.616,
      "step": 2000
    },
    {
      "epoch": 1.966041108132261,
      "grad_norm": 0.2072182446718216,
      "learning_rate": 6.921770545784254e-05,
      "loss": 0.1513,
      "step": 2200
    },
    {
      "epoch": 1.966041108132261,
      "eval_loss": 0.16549332439899445,
      "eval_runtime": 56.2064,
      "eval_samples_per_second": 3.22,
      "eval_steps_per_second": 1.619,
      "step": 2200
    },
    {
      "epoch": 2.1447721179624666,
      "grad_norm": 0.1624869555234909,
      "learning_rate": 6.375682535204167e-05,
      "loss": 0.1338,
      "step": 2400
    },
    {
      "epoch": 2.1447721179624666,
      "eval_loss": 0.1687156856060028,
      "eval_runtime": 56.2284,
      "eval_samples_per_second": 3.219,
      "eval_steps_per_second": 1.618,
      "step": 2400
    }
  ],
  "logging_steps": 200,
  "max_steps": 5595,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6581699483183104e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}