{
  "best_metric": 0.72297174,
  "best_model_checkpoint": "Qwen2.5-3B-Instruct_lora/checkpoint-128",
  "epoch": 0.997078870496592,
  "eval_steps": 100,
  "global_step": 128,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007789678675754625,
      "grad_norm": 0.4715007245540619,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.8726969361305237,
      "memory(GiB)": 78.63,
      "step": 1,
      "token_acc": 0.7654501216545012,
      "train_speed(iter/s)": 0.019242
    },
    {
      "epoch": 0.03894839337877313,
      "grad_norm": 0.4972277879714966,
      "learning_rate": 7.142857142857143e-05,
      "loss": 1.0096402168273926,
      "memory(GiB)": 78.63,
      "step": 5,
      "token_acc": 0.7308917197452229,
      "train_speed(iter/s)": 0.022704
    },
    {
      "epoch": 0.07789678675754626,
      "grad_norm": 0.28860655426979065,
      "learning_rate": 9.984840253435568e-05,
      "loss": 0.8709381103515625,
      "memory(GiB)": 78.63,
      "step": 10,
      "token_acc": 0.7596180459451979,
      "train_speed(iter/s)": 0.023588
    },
    {
      "epoch": 0.11684518013631938,
      "grad_norm": 0.25548580288887024,
      "learning_rate": 9.892530053338909e-05,
      "loss": 0.8737223625183106,
      "memory(GiB)": 78.63,
      "step": 15,
      "token_acc": 0.7533364421646348,
      "train_speed(iter/s)": 0.023387
    },
    {
      "epoch": 0.15579357351509251,
      "grad_norm": 0.19688600301742554,
      "learning_rate": 9.71788333255991e-05,
      "loss": 0.8075315475463867,
      "memory(GiB)": 78.63,
      "step": 20,
      "token_acc": 0.7687880916238077,
      "train_speed(iter/s)": 0.023721
    },
    {
      "epoch": 0.19474196689386564,
      "grad_norm": 0.2052084058523178,
      "learning_rate": 9.46383922548932e-05,
      "loss": 0.8128433227539062,
      "memory(GiB)": 78.63,
      "step": 25,
      "token_acc": 0.7695909532644226,
      "train_speed(iter/s)": 0.023782
    },
    {
      "epoch": 0.23369036027263876,
      "grad_norm": 0.1965310275554657,
      "learning_rate": 9.134673047280645e-05,
      "loss": 0.8438224792480469,
      "memory(GiB)": 78.63,
      "step": 30,
      "token_acc": 0.7579175899192014,
      "train_speed(iter/s)": 0.023976
    },
    {
      "epoch": 0.2726387536514119,
      "grad_norm": 0.17103135585784912,
      "learning_rate": 8.735924344455732e-05,
      "loss": 0.7569032192230225,
      "memory(GiB)": 78.63,
      "step": 35,
      "token_acc": 0.7819433973797347,
      "train_speed(iter/s)": 0.024023
    },
    {
      "epoch": 0.31158714703018503,
      "grad_norm": 0.2070111483335495,
      "learning_rate": 8.274303669726426e-05,
      "loss": 0.8125999450683594,
      "memory(GiB)": 78.63,
      "step": 40,
      "token_acc": 0.7662681276988743,
      "train_speed(iter/s)": 0.024049
    },
    {
      "epoch": 0.3505355404089581,
      "grad_norm": 0.22748339176177979,
      "learning_rate": 7.757579649921354e-05,
      "loss": 0.8048953056335449,
      "memory(GiB)": 78.63,
      "step": 45,
      "token_acc": 0.7660392978013684,
      "train_speed(iter/s)": 0.023996
    },
    {
      "epoch": 0.3894839337877313,
      "grad_norm": 0.19850845634937286,
      "learning_rate": 7.19444824755478e-05,
      "loss": 0.7799190044403076,
      "memory(GiB)": 78.63,
      "step": 50,
      "token_acc": 0.7750339878312421,
      "train_speed(iter/s)": 0.024045
    },
    {
      "epoch": 0.42843232716650437,
      "grad_norm": 0.21935951709747314,
      "learning_rate": 6.594386416238094e-05,
      "loss": 0.777038288116455,
      "memory(GiB)": 78.63,
      "step": 55,
      "token_acc": 0.7741007536314576,
      "train_speed(iter/s)": 0.023874
    },
    {
      "epoch": 0.4673807205452775,
      "grad_norm": 0.20794467628002167,
      "learning_rate": 5.967492612770999e-05,
      "loss": 0.7499256134033203,
      "memory(GiB)": 78.63,
      "step": 60,
      "token_acc": 0.7789702908694975,
      "train_speed(iter/s)": 0.023757
    },
    {
      "epoch": 0.5063291139240507,
      "grad_norm": 0.20109890401363373,
      "learning_rate": 5.324316849938715e-05,
      "loss": 0.7714523792266845,
      "memory(GiB)": 78.63,
      "step": 65,
      "token_acc": 0.7740315796428021,
      "train_speed(iter/s)": 0.018807
    },
    {
      "epoch": 0.5452775073028238,
      "grad_norm": 0.21935297548770905,
      "learning_rate": 4.675683150061285e-05,
      "loss": 0.748004674911499,
      "memory(GiB)": 78.63,
      "step": 70,
      "token_acc": 0.7820459013496499,
      "train_speed(iter/s)": 0.018099
    },
    {
      "epoch": 0.5842259006815969,
      "grad_norm": 0.2540643811225891,
      "learning_rate": 4.032507387229002e-05,
      "loss": 0.7518786430358887,
      "memory(GiB)": 78.63,
      "step": 75,
      "token_acc": 0.7795054192401047,
      "train_speed(iter/s)": 0.01846
    },
    {
      "epoch": 0.6231742940603701,
      "grad_norm": 0.20665927231311798,
      "learning_rate": 3.4056135837619074e-05,
      "loss": 0.7919217586517334,
      "memory(GiB)": 78.63,
      "step": 80,
      "token_acc": 0.7714305789769533,
      "train_speed(iter/s)": 0.01882
    },
    {
      "epoch": 0.6621226874391432,
      "grad_norm": 0.21973538398742676,
      "learning_rate": 2.805551752445222e-05,
      "loss": 0.735324478149414,
      "memory(GiB)": 78.63,
      "step": 85,
      "token_acc": 0.7844363538897947,
      "train_speed(iter/s)": 0.019084
    },
    {
      "epoch": 0.7010710808179162,
      "grad_norm": 0.22932057082653046,
      "learning_rate": 2.2424203500786474e-05,
      "loss": 0.7479868412017823,
      "memory(GiB)": 78.63,
      "step": 90,
      "token_acc": 0.7778178864723401,
      "train_speed(iter/s)": 0.01932
    },
    {
      "epoch": 0.7400194741966893,
      "grad_norm": 0.2437477707862854,
      "learning_rate": 1.725696330273575e-05,
      "loss": 0.784179973602295,
      "memory(GiB)": 78.63,
      "step": 95,
      "token_acc": 0.7689784644598852,
      "train_speed(iter/s)": 0.019591
    },
    {
      "epoch": 0.7789678675754625,
      "grad_norm": 0.26622432470321655,
      "learning_rate": 1.2640756555442684e-05,
      "loss": 0.7635871410369873,
      "memory(GiB)": 78.63,
      "step": 100,
      "token_acc": 0.7789733040981939,
      "train_speed(iter/s)": 0.019811
    },
    {
      "epoch": 0.7789678675754625,
      "eval_loss": 0.7245169281959534,
      "eval_runtime": 1.7166,
      "eval_samples_per_second": 2.33,
      "eval_steps_per_second": 1.165,
      "step": 100
    },
    {
      "epoch": 0.8179162609542356,
      "grad_norm": 0.2450593113899231,
      "learning_rate": 8.65326952719357e-06,
      "loss": 0.7425096988677978,
      "memory(GiB)": 78.63,
      "step": 105,
      "token_acc": 0.7821539141972579,
      "train_speed(iter/s)": 0.020023
    },
    {
      "epoch": 0.8568646543330087,
      "grad_norm": 0.25075599551200867,
      "learning_rate": 5.361607745106817e-06,
      "loss": 0.7099203586578369,
      "memory(GiB)": 78.63,
      "step": 110,
      "token_acc": 0.7904701127419698,
      "train_speed(iter/s)": 0.015457
    },
    {
      "epoch": 0.8958130477117819,
      "grad_norm": 0.2313348650932312,
      "learning_rate": 2.8211666744009047e-06,
      "loss": 0.7207077026367188,
      "memory(GiB)": 78.63,
      "step": 115,
      "token_acc": 0.7838124464100366,
      "train_speed(iter/s)": 0.01571
    },
    {
      "epoch": 0.934761441090555,
      "grad_norm": 0.24124138057231903,
      "learning_rate": 1.0746994666109234e-06,
      "loss": 0.7263786315917968,
      "memory(GiB)": 78.63,
      "step": 120,
      "token_acc": 0.7858320369639415,
      "train_speed(iter/s)": 0.015968
    },
    {
      "epoch": 0.9737098344693281,
      "grad_norm": 0.24106207489967346,
      "learning_rate": 1.5159746564433198e-07,
      "loss": 0.7232465744018555,
      "memory(GiB)": 78.63,
      "step": 125,
      "token_acc": 0.7840324763193505,
      "train_speed(iter/s)": 0.016211
    },
    {
      "epoch": 0.997078870496592,
      "eval_loss": 0.7229717373847961,
      "eval_runtime": 1.6035,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 1.247,
      "step": 128
    }
  ],
  "logging_steps": 5,
  "max_steps": 128,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 3.569669711559721e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}