{
  "best_global_step": 400,
  "best_metric": 1.5033758878707886,
  "best_model_checkpoint": "./Llama-3.2-4b-ru-lora/checkpoint-400",
  "epoch": 2.8806584362139915,
  "eval_steps": 100,
  "global_step": 700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.102880658436214,
      "grad_norm": 0.8221918344497681,
      "learning_rate": 8e-05,
      "loss": 2.379297180175781,
      "step": 25
    },
    {
      "epoch": 0.205761316872428,
      "grad_norm": 0.8748275637626648,
      "learning_rate": 9.728183118741059e-05,
      "loss": 1.9784544372558595,
      "step": 50
    },
    {
      "epoch": 0.30864197530864196,
      "grad_norm": 0.750031590461731,
      "learning_rate": 9.370529327610873e-05,
      "loss": 1.8810975646972656,
      "step": 75
    },
    {
      "epoch": 0.411522633744856,
      "grad_norm": 0.9241881966590881,
      "learning_rate": 9.012875536480687e-05,
      "loss": 1.8589222717285157,
      "step": 100
    },
    {
      "epoch": 0.411522633744856,
      "eval_loss": 1.582080602645874,
      "eval_runtime": 22.7091,
      "eval_samples_per_second": 4.536,
      "eval_steps_per_second": 4.536,
      "step": 100
    },
    {
      "epoch": 0.51440329218107,
      "grad_norm": 0.955057680606842,
      "learning_rate": 8.655221745350501e-05,
      "loss": 1.762770538330078,
      "step": 125
    },
    {
      "epoch": 0.6172839506172839,
      "grad_norm": 0.8256515860557556,
      "learning_rate": 8.297567954220316e-05,
      "loss": 1.7971592712402344,
      "step": 150
    },
    {
      "epoch": 0.720164609053498,
      "grad_norm": 0.8769187331199646,
      "learning_rate": 7.939914163090129e-05,
      "loss": 1.8236392211914063,
      "step": 175
    },
    {
      "epoch": 0.823045267489712,
      "grad_norm": 0.8237674832344055,
      "learning_rate": 7.582260371959943e-05,
      "loss": 1.8212730407714843,
      "step": 200
    },
    {
      "epoch": 0.823045267489712,
      "eval_loss": 1.5252765417099,
      "eval_runtime": 22.7366,
      "eval_samples_per_second": 4.53,
      "eval_steps_per_second": 4.53,
      "step": 200
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 0.8813307881355286,
      "learning_rate": 7.224606580829757e-05,
      "loss": 1.7407562255859375,
      "step": 225
    },
    {
      "epoch": 1.02880658436214,
      "grad_norm": 0.7620531320571899,
      "learning_rate": 6.86695278969957e-05,
      "loss": 1.667363739013672,
      "step": 250
    },
    {
      "epoch": 1.131687242798354,
      "grad_norm": 0.8878408670425415,
      "learning_rate": 6.509298998569385e-05,
      "loss": 1.4600253295898438,
      "step": 275
    },
    {
      "epoch": 1.2345679012345678,
      "grad_norm": 0.9481920003890991,
      "learning_rate": 6.1516452074392e-05,
      "loss": 1.4149290466308593,
      "step": 300
    },
    {
      "epoch": 1.2345679012345678,
      "eval_loss": 1.5132408142089844,
      "eval_runtime": 22.7071,
      "eval_samples_per_second": 4.536,
      "eval_steps_per_second": 4.536,
      "step": 300
    },
    {
      "epoch": 1.337448559670782,
      "grad_norm": 1.0244613885879517,
      "learning_rate": 5.793991416309014e-05,
      "loss": 1.4301467895507813,
      "step": 325
    },
    {
      "epoch": 1.4403292181069958,
      "grad_norm": 1.1760555505752563,
      "learning_rate": 5.4363376251788265e-05,
      "loss": 1.4518887329101562,
      "step": 350
    },
    {
      "epoch": 1.5432098765432098,
      "grad_norm": 1.2840839624404907,
      "learning_rate": 5.078683834048641e-05,
      "loss": 1.4007278442382813,
      "step": 375
    },
    {
      "epoch": 1.646090534979424,
      "grad_norm": 0.9614652395248413,
      "learning_rate": 4.721030042918455e-05,
      "loss": 1.4439279174804687,
      "step": 400
    },
    {
      "epoch": 1.646090534979424,
      "eval_loss": 1.5033758878707886,
      "eval_runtime": 22.7479,
      "eval_samples_per_second": 4.528,
      "eval_steps_per_second": 4.528,
      "step": 400
    },
    {
      "epoch": 1.7489711934156378,
      "grad_norm": 1.3173755407333374,
      "learning_rate": 4.363376251788269e-05,
      "loss": 1.3896630859375,
      "step": 425
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 1.3660273551940918,
      "learning_rate": 4.005722460658083e-05,
      "loss": 1.4115650939941407,
      "step": 450
    },
    {
      "epoch": 1.954732510288066,
      "grad_norm": 1.2740566730499268,
      "learning_rate": 3.648068669527897e-05,
      "loss": 1.4732164001464845,
      "step": 475
    },
    {
      "epoch": 2.05761316872428,
      "grad_norm": 1.1533112525939941,
      "learning_rate": 3.290414878397711e-05,
      "loss": 1.2941168212890626,
      "step": 500
    },
    {
      "epoch": 2.05761316872428,
      "eval_loss": 1.5046026706695557,
      "eval_runtime": 22.7013,
      "eval_samples_per_second": 4.537,
      "eval_steps_per_second": 4.537,
      "step": 500
    },
    {
      "epoch": 2.1604938271604937,
      "grad_norm": 1.2710565328598022,
      "learning_rate": 2.9327610872675255e-05,
      "loss": 1.052796401977539,
      "step": 525
    },
    {
      "epoch": 2.263374485596708,
      "grad_norm": 1.48249089717865,
      "learning_rate": 2.575107296137339e-05,
      "loss": 1.1035122680664062,
      "step": 550
    },
    {
      "epoch": 2.366255144032922,
      "grad_norm": 1.1420801877975464,
      "learning_rate": 2.217453505007153e-05,
      "loss": 1.1403243255615234,
      "step": 575
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 1.2909879684448242,
      "learning_rate": 1.859799713876967e-05,
      "loss": 1.0371189880371094,
      "step": 600
    },
    {
      "epoch": 2.4691358024691357,
      "eval_loss": 1.5498303174972534,
      "eval_runtime": 22.7264,
      "eval_samples_per_second": 4.532,
      "eval_steps_per_second": 4.532,
      "step": 600
    },
    {
      "epoch": 2.57201646090535,
      "grad_norm": 1.2476613521575928,
      "learning_rate": 1.5021459227467813e-05,
      "loss": 1.0042486572265625,
      "step": 625
    },
    {
      "epoch": 2.674897119341564,
      "grad_norm": 1.3221091032028198,
      "learning_rate": 1.1444921316165953e-05,
      "loss": 1.0447573089599609,
      "step": 650
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 1.1615115404129028,
      "learning_rate": 7.868383404864092e-06,
      "loss": 1.1024551391601562,
      "step": 675
    },
    {
      "epoch": 2.8806584362139915,
      "grad_norm": 1.4701781272888184,
      "learning_rate": 4.2918454935622316e-06,
      "loss": 1.0973941040039064,
      "step": 700
    },
    {
      "epoch": 2.8806584362139915,
      "eval_loss": 1.5574243068695068,
      "eval_runtime": 22.7828,
      "eval_samples_per_second": 4.521,
      "eval_steps_per_second": 4.521,
      "step": 700
    }
  ],
  "logging_steps": 25,
  "max_steps": 729,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.86041359785984e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}