{
  "best_global_step": 142,
  "best_metric": 0.22699420154094696,
  "best_model_checkpoint": "/content/DualMind100_agentic/checkpoint-142",
  "epoch": 2.0,
  "eval_steps": 50,
  "global_step": 142,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.16182193332642783,
      "epoch": 0.07092198581560284,
      "grad_norm": 0.5555469989776611,
      "learning_rate": 5e-05,
      "loss": 0.43506731986999514,
      "mean_token_accuracy": 0.8946891106665135,
      "num_tokens": 123430.0,
      "step": 5
    },
    {
      "entropy": 0.26333686951547863,
      "epoch": 0.14184397163120568,
      "grad_norm": 0.2982931435108185,
      "learning_rate": 9.99862592554908e-05,
      "loss": 0.3709733247756958,
      "mean_token_accuracy": 0.898155527561903,
      "num_tokens": 240929.0,
      "step": 10
    },
    {
      "entropy": 0.34238293119706215,
      "epoch": 0.2127659574468085,
      "grad_norm": 0.27323758602142334,
      "learning_rate": 9.950612572673255e-05,
      "loss": 0.3364511489868164,
      "mean_token_accuracy": 0.906800027191639,
      "num_tokens": 373968.0,
      "step": 15
    },
    {
      "entropy": 0.2675066887401044,
      "epoch": 0.28368794326241137,
      "grad_norm": 0.21929873526096344,
      "learning_rate": 9.834648865751254e-05,
      "loss": 0.2727418184280396,
      "mean_token_accuracy": 0.9184959597885609,
      "num_tokens": 494846.0,
      "step": 20
    },
    {
      "entropy": 0.2389053585124202,
      "epoch": 0.3546099290780142,
      "grad_norm": 0.19836418330669403,
      "learning_rate": 9.652326481535435e-05,
      "loss": 0.3025625705718994,
      "mean_token_accuracy": 0.9214970305562019,
      "num_tokens": 630306.0,
      "step": 25
    },
    {
      "entropy": 0.25811560889706014,
      "epoch": 0.425531914893617,
      "grad_norm": 0.19982172548770905,
      "learning_rate": 9.406147912470143e-05,
      "loss": 0.330981969833374,
      "mean_token_accuracy": 0.9110702231526375,
      "num_tokens": 758739.0,
      "step": 30
    },
    {
      "entropy": 0.24841026657959447,
      "epoch": 0.49645390070921985,
      "grad_norm": 0.16917800903320312,
      "learning_rate": 9.099492118367123e-05,
      "loss": 0.26140174865722654,
      "mean_token_accuracy": 0.9228238843381404,
      "num_tokens": 886920.0,
      "step": 35
    },
    {
      "entropy": 0.2569453418778721,
      "epoch": 0.5673758865248227,
      "grad_norm": 0.2460022270679474,
      "learning_rate": 8.736568148000386e-05,
      "loss": 0.28326830863952634,
      "mean_token_accuracy": 0.9195088751614093,
      "num_tokens": 1010775.0,
      "step": 40
    },
    {
      "entropy": 0.29926245249807837,
      "epoch": 0.6382978723404256,
      "grad_norm": 0.17467764019966125,
      "learning_rate": 8.322357367194109e-05,
      "loss": 0.3357896566390991,
      "mean_token_accuracy": 0.9085973531007767,
      "num_tokens": 1134373.0,
      "step": 45
    },
    {
      "entropy": 0.2604390908149071,
      "epoch": 0.7092198581560284,
      "grad_norm": 0.16242940723896027,
      "learning_rate": 7.86254508636036e-05,
      "loss": 0.2815252304077148,
      "mean_token_accuracy": 0.9152372993528843,
      "num_tokens": 1256155.0,
      "step": 50
    },
    {
      "epoch": 0.7092198581560284,
      "eval_entropy": 0.22815538255963475,
      "eval_loss": 0.23423044383525848,
      "eval_mean_token_accuracy": 0.9300008341670036,
      "eval_num_tokens": 1256155.0,
      "eval_runtime": 5.3417,
      "eval_samples_per_second": 8.986,
      "eval_steps_per_second": 8.986,
      "step": 50
    },
    {
      "entropy": 0.2784213587117847,
      "epoch": 0.7801418439716312,
      "grad_norm": 0.1878730058670044,
      "learning_rate": 7.363442525942826e-05,
      "loss": 0.2845616340637207,
      "mean_token_accuracy": 0.9107947684824467,
      "num_tokens": 1378868.0,
      "step": 55
    },
    {
      "entropy": 0.29253384250914677,
      "epoch": 0.851063829787234,
      "grad_norm": 0.17584209144115448,
      "learning_rate": 6.831900190841232e-05,
      "loss": 0.31951193809509276,
      "mean_token_accuracy": 0.9116400554776192,
      "num_tokens": 1509648.0,
      "step": 60
    },
    {
      "entropy": 0.27282086764462293,
      "epoch": 0.9219858156028369,
      "grad_norm": 0.1782023161649704,
      "learning_rate": 6.275213842808383e-05,
      "loss": 0.28685288429260253,
      "mean_token_accuracy": 0.9180420659482479,
      "num_tokens": 1640179.0,
      "step": 65
    },
    {
      "entropy": 0.26427618057350627,
      "epoch": 0.9929078014184397,
      "grad_norm": 0.2283456027507782,
      "learning_rate": 5.70102436140943e-05,
      "loss": 0.29142701625823975,
      "mean_token_accuracy": 0.9174253843724728,
      "num_tokens": 1764592.0,
      "step": 70
    },
    {
      "entropy": 0.2524041742629682,
      "epoch": 1.0567375886524824,
      "grad_norm": 0.16881126165390015,
      "learning_rate": 5.117212868016303e-05,
      "loss": 0.276317834854126,
      "mean_token_accuracy": 0.9248060741358333,
      "num_tokens": 1874267.0,
      "step": 75
    },
    {
      "entropy": 0.29945559421321377,
      "epoch": 1.127659574468085,
      "grad_norm": 0.2186647355556488,
      "learning_rate": 4.5317925523282464e-05,
      "loss": 0.29511370658874514,
      "mean_token_accuracy": 0.907706456631422,
      "num_tokens": 1991985.0,
      "step": 80
    },
    {
      "entropy": 0.24618784408958164,
      "epoch": 1.198581560283688,
      "grad_norm": 0.15828007459640503,
      "learning_rate": 3.952798686169279e-05,
      "loss": 0.2731397867202759,
      "mean_token_accuracy": 0.9240930497646331,
      "num_tokens": 2118168.0,
      "step": 85
    },
    {
      "entropy": 0.2373207072203513,
      "epoch": 1.2695035460992907,
      "grad_norm": 0.20958004891872406,
      "learning_rate": 3.388178334194232e-05,
      "loss": 0.27325007915496824,
      "mean_token_accuracy": 0.9249208129942417,
      "num_tokens": 2240951.0,
      "step": 90
    },
    {
      "entropy": 0.22738294367445633,
      "epoch": 1.3404255319148937,
      "grad_norm": 0.19565178453922272,
      "learning_rate": 2.8456812752951485e-05,
      "loss": 0.25463979244232177,
      "mean_token_accuracy": 0.9295149140059948,
      "num_tokens": 2364856.0,
      "step": 95
    },
    {
      "entropy": 0.247153190523386,
      "epoch": 1.4113475177304964,
      "grad_norm": 0.1792992502450943,
      "learning_rate": 2.3327536318821495e-05,
      "loss": 0.2785398244857788,
      "mean_token_accuracy": 0.9210605584084988,
      "num_tokens": 2496977.0,
      "step": 100
    },
    {
      "epoch": 1.4113475177304964,
      "eval_entropy": 0.2170733372719648,
      "eval_loss": 0.22858719527721405,
      "eval_mean_token_accuracy": 0.9310164228081703,
      "eval_num_tokens": 2496977.0,
      "eval_runtime": 5.44,
      "eval_samples_per_second": 8.824,
      "eval_steps_per_second": 8.824,
      "step": 100
    },
    {
      "entropy": 0.2818379683652893,
      "epoch": 1.4822695035460993,
      "grad_norm": 0.17297224700450897,
      "learning_rate": 1.8564356670455767e-05,
      "loss": 0.28323366641998293,
      "mean_token_accuracy": 0.9150983460247517,
      "num_tokens": 2625956.0,
      "step": 105
    },
    {
      "entropy": 0.25849783700541595,
      "epoch": 1.5531914893617023,
      "grad_norm": 0.17231720685958862,
      "learning_rate": 1.4232651523993634e-05,
      "loss": 0.26291708946228026,
      "mean_token_accuracy": 0.9214990958571434,
      "num_tokens": 2762411.0,
      "step": 110
    },
    {
      "entropy": 0.23584751789458097,
      "epoch": 1.624113475177305,
      "grad_norm": 0.19976992905139923,
      "learning_rate": 1.0391876329443533e-05,
      "loss": 0.24894940853118896,
      "mean_token_accuracy": 0.9259149916470051,
      "num_tokens": 2887514.0,
      "step": 115
    },
    {
      "entropy": 0.2430717541486956,
      "epoch": 1.6950354609929077,
      "grad_norm": 0.21702121198177338,
      "learning_rate": 7.094748206242796e-06,
      "loss": 0.2538208246231079,
      "mean_token_accuracy": 0.9232472732663155,
      "num_tokens": 3013243.0,
      "step": 120
    },
    {
      "entropy": 0.23513522319262847,
      "epoch": 1.7659574468085106,
      "grad_norm": 0.16074544191360474,
      "learning_rate": 4.386522366755169e-06,
      "loss": 0.22128725051879883,
      "mean_token_accuracy": 0.9245993517339229,
      "num_tokens": 3134144.0,
      "step": 125
    },
    {
      "entropy": 0.24327249150956048,
      "epoch": 1.8368794326241136,
      "grad_norm": 0.16139043867588043,
      "learning_rate": 2.3043709592610485e-06,
      "loss": 0.26496360301971433,
      "mean_token_accuracy": 0.9262310199439525,
      "num_tokens": 3260737.0,
      "step": 130
    },
    {
      "entropy": 0.23524304252350703,
      "epoch": 1.9078014184397163,
      "grad_norm": 0.16922682523727417,
      "learning_rate": 8.768728562211947e-07,
      "loss": 0.2534662961959839,
      "mean_token_accuracy": 0.924710838496685,
      "num_tokens": 3383649.0,
      "step": 135
    },
    {
      "entropy": 0.2577046745980624,
      "epoch": 1.978723404255319,
      "grad_norm": 0.18219797313213348,
      "learning_rate": 1.236213907994943e-07,
      "loss": 0.2727572679519653,
      "mean_token_accuracy": 0.9200983986258506,
      "num_tokens": 3511363.0,
      "step": 140
    },
    {
      "epoch": 2.0,
      "eval_entropy": 0.21362574538215995,
      "eval_loss": 0.22699420154094696,
      "eval_mean_token_accuracy": 0.9318861390153567,
      "eval_num_tokens": 3550882.0,
      "eval_runtime": 5.4445,
      "eval_samples_per_second": 8.816,
      "eval_steps_per_second": 8.816,
      "step": 142
    },
    {
      "epoch": 2.0,
      "step": 142,
      "total_flos": 3.08394581409792e+16,
      "train_loss": 0.2892465377357644,
      "train_runtime": 1218.7991,
      "train_samples_per_second": 1.851,
      "train_steps_per_second": 0.117
    }
  ],
  "logging_steps": 5,
  "max_steps": 142,
  "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": 3.08394581409792e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}