{
  "best_global_step": 900,
  "best_metric": 0.595556914806366,
  "best_model_checkpoint": "/content/drive/MyDrive/Colab Notebooks/AITF_baseqwen3_8B_V1.2/checkpoint-900",
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 929,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.021531422419593594,
      "grad_norm": 0.38749924302101135,
      "learning_rate": 7.6e-05,
      "loss": 1.7456,
      "step": 20
    },
    {
      "epoch": 0.04306284483918719,
      "grad_norm": 0.23645007610321045,
      "learning_rate": 0.00015600000000000002,
      "loss": 1.2508,
      "step": 40
    },
    {
      "epoch": 0.06459426725878079,
      "grad_norm": 0.16630907356739044,
      "learning_rate": 0.0001999482703462211,
      "loss": 1.0123,
      "step": 60
    },
    {
      "epoch": 0.08612568967837438,
      "grad_norm": 0.1874742954969406,
      "learning_rate": 0.00019946334007549978,
      "loss": 0.9242,
      "step": 80
    },
    {
      "epoch": 0.10765711209796797,
      "grad_norm": 0.1982610672712326,
      "learning_rate": 0.0001984704140331751,
      "loss": 0.8479,
      "step": 100
    },
    {
      "epoch": 0.10765711209796797,
      "eval_loss": 0.7989935874938965,
      "eval_runtime": 368.247,
      "eval_samples_per_second": 10.105,
      "eval_steps_per_second": 2.528,
      "step": 100
    },
    {
      "epoch": 0.12918853451756157,
      "grad_norm": 0.2351531982421875,
      "learning_rate": 0.0001969745634568572,
      "loss": 0.8075,
      "step": 120
    },
    {
      "epoch": 0.15071995693715515,
      "grad_norm": 0.2675715982913971,
      "learning_rate": 0.00019498342820427794,
      "loss": 0.8,
      "step": 140
    },
    {
      "epoch": 0.17225137935674875,
      "grad_norm": 0.2564864754676819,
      "learning_rate": 0.00019250717773373462,
      "loss": 0.7552,
      "step": 160
    },
    {
      "epoch": 0.19378280177634236,
      "grad_norm": 0.28067538142204285,
      "learning_rate": 0.0001895584591649349,
      "loss": 0.7529,
      "step": 180
    },
    {
      "epoch": 0.21531422419593593,
      "grad_norm": 0.25266703963279724,
      "learning_rate": 0.00018615233268551643,
      "loss": 0.727,
      "step": 200
    },
    {
      "epoch": 0.21531422419593593,
      "eval_loss": 0.7019057869911194,
      "eval_runtime": 367.354,
      "eval_samples_per_second": 10.129,
      "eval_steps_per_second": 2.534,
      "step": 200
    },
    {
      "epoch": 0.23684564661552954,
      "grad_norm": 0.2786823809146881,
      "learning_rate": 0.00018230619463314266,
      "loss": 0.7363,
      "step": 220
    },
    {
      "epoch": 0.25837706903512314,
      "grad_norm": 0.2877826690673828,
      "learning_rate": 0.0001780396886460237,
      "loss": 0.7094,
      "step": 240
    },
    {
      "epoch": 0.27990849145471675,
      "grad_norm": 0.3062780499458313,
      "learning_rate": 0.00017337460533564845,
      "loss": 0.7013,
      "step": 260
    },
    {
      "epoch": 0.3014399138743103,
      "grad_norm": 0.29140764474868774,
      "learning_rate": 0.0001683347709941367,
      "loss": 0.6772,
      "step": 280
    },
    {
      "epoch": 0.3229713362939039,
      "grad_norm": 0.2650732398033142,
      "learning_rate": 0.00016294592590462316,
      "loss": 0.6896,
      "step": 300
    },
    {
      "epoch": 0.3229713362939039,
      "eval_loss": 0.6610061526298523,
      "eval_runtime": 367.3279,
      "eval_samples_per_second": 10.13,
      "eval_steps_per_second": 2.535,
      "step": 300
    },
    {
      "epoch": 0.3445027587134975,
      "grad_norm": 0.2866383194923401,
      "learning_rate": 0.00015723559287618728,
      "loss": 0.6685,
      "step": 320
    },
    {
      "epoch": 0.3660341811330911,
      "grad_norm": 0.2941460609436035,
      "learning_rate": 0.00015123293667476887,
      "loss": 0.6743,
      "step": 340
    },
    {
      "epoch": 0.3875656035526847,
      "grad_norm": 0.2664410471916199,
      "learning_rate": 0.00014496861506800758,
      "loss": 0.6755,
      "step": 360
    },
    {
      "epoch": 0.4090970259722783,
      "grad_norm": 0.2797100245952606,
      "learning_rate": 0.00013847462224477538,
      "loss": 0.6691,
      "step": 380
    },
    {
      "epoch": 0.43062844839187187,
      "grad_norm": 0.30962368845939636,
      "learning_rate": 0.00013178412540911457,
      "loss": 0.6658,
      "step": 400
    },
    {
      "epoch": 0.43062844839187187,
      "eval_loss": 0.6373162865638733,
      "eval_runtime": 367.4733,
      "eval_samples_per_second": 10.126,
      "eval_steps_per_second": 2.534,
      "step": 400
    },
    {
      "epoch": 0.45215987081146547,
      "grad_norm": 0.29752910137176514,
      "learning_rate": 0.00012493129538315788,
      "loss": 0.6587,
      "step": 420
    },
    {
      "epoch": 0.4736912932310591,
      "grad_norm": 0.304659903049469,
      "learning_rate": 0.00011795113208420208,
      "loss": 0.6466,
      "step": 440
    },
    {
      "epoch": 0.4952227156506527,
      "grad_norm": 0.3261611759662628,
      "learning_rate": 0.00011087928576728865,
      "loss": 0.6216,
      "step": 460
    },
    {
      "epoch": 0.5167541380702463,
      "grad_norm": 0.3342382311820984,
      "learning_rate": 0.00010375187494627098,
      "loss": 0.6435,
      "step": 480
    },
    {
      "epoch": 0.5382855604898399,
      "grad_norm": 0.2871527075767517,
      "learning_rate": 9.660530192331191e-05,
      "loss": 0.643,
      "step": 500
    },
    {
      "epoch": 0.5382855604898399,
      "eval_loss": 0.6207689642906189,
      "eval_runtime": 367.6602,
      "eval_samples_per_second": 10.121,
      "eval_steps_per_second": 2.532,
      "step": 500
    },
    {
      "epoch": 0.5598169829094335,
      "grad_norm": 0.3142107129096985,
      "learning_rate": 8.947606686897045e-05,
      "loss": 0.6355,
      "step": 520
    },
    {
      "epoch": 0.5813484053290271,
      "grad_norm": 0.3034207820892334,
      "learning_rate": 8.240058140243834e-05,
      "loss": 0.639,
      "step": 540
    },
    {
      "epoch": 0.6028798277486206,
      "grad_norm": 0.2917335033416748,
      "learning_rate": 7.541498262404125e-05,
      "loss": 0.6345,
      "step": 560
    },
    {
      "epoch": 0.6244112501682142,
      "grad_norm": 0.301921010017395,
      "learning_rate": 6.855494854980857e-05,
      "loss": 0.6361,
      "step": 580
    },
    {
      "epoch": 0.6459426725878078,
      "grad_norm": 0.2999766170978546,
      "learning_rate": 6.185551589075482e-05,
      "loss": 0.6266,
      "step": 600
    },
    {
      "epoch": 0.6459426725878078,
      "eval_loss": 0.6095116138458252,
      "eval_runtime": 369.6908,
      "eval_samples_per_second": 10.065,
      "eval_steps_per_second": 2.518,
      "step": 600
    },
    {
      "epoch": 0.6674740950074014,
      "grad_norm": 0.2857365310192108,
      "learning_rate": 5.535090110754131e-05,
      "loss": 0.6379,
      "step": 620
    },
    {
      "epoch": 0.689005517426995,
      "grad_norm": 0.2973370850086212,
      "learning_rate": 4.9074325654457446e-05,
      "loss": 0.6126,
      "step": 640
    },
    {
      "epoch": 0.7105369398465886,
      "grad_norm": 0.3231595754623413,
      "learning_rate": 4.305784630526416e-05,
      "loss": 0.6105,
      "step": 660
    },
    {
      "epoch": 0.7320683622661822,
      "grad_norm": 0.2923620939254761,
      "learning_rate": 3.7332191427488784e-05,
      "loss": 0.6266,
      "step": 680
    },
    {
      "epoch": 0.7535997846857758,
      "grad_norm": 0.32130059599876404,
      "learning_rate": 3.192660404137729e-05,
      "loss": 0.6088,
      "step": 700
    },
    {
      "epoch": 0.7535997846857758,
      "eval_loss": 0.6016086935997009,
      "eval_runtime": 367.7262,
      "eval_samples_per_second": 10.119,
      "eval_steps_per_second": 2.532,
      "step": 700
    },
    {
      "epoch": 0.7751312071053694,
      "grad_norm": 0.2963238060474396,
      "learning_rate": 2.6868692465060828e-05,
      "loss": 0.6256,
      "step": 720
    },
    {
      "epoch": 0.796662629524963,
      "grad_norm": 0.2954428195953369,
      "learning_rate": 2.2184289308744844e-05,
      "loss": 0.6201,
      "step": 740
    },
    {
      "epoch": 0.8181940519445566,
      "grad_norm": 0.3122366964817047,
      "learning_rate": 1.7897319538090962e-05,
      "loss": 0.6164,
      "step": 760
    },
    {
      "epoch": 0.8397254743641501,
      "grad_norm": 0.3299662470817566,
      "learning_rate": 1.402967828063897e-05,
      "loss": 0.6026,
      "step": 780
    },
    {
      "epoch": 0.8612568967837437,
      "grad_norm": 0.2979361414909363,
      "learning_rate": 1.0601118999356907e-05,
      "loss": 0.6103,
      "step": 800
    },
    {
      "epoch": 0.8612568967837437,
      "eval_loss": 0.596613347530365,
      "eval_runtime": 367.8439,
      "eval_samples_per_second": 10.116,
      "eval_steps_per_second": 2.531,
      "step": 800
    },
    {
      "epoch": 0.8827883192033373,
      "grad_norm": 0.3615986108779907,
      "learning_rate": 7.629152604458156e-06,
      "loss": 0.6052,
      "step": 820
    },
    {
      "epoch": 0.9043197416229309,
      "grad_norm": 0.31248706579208374,
      "learning_rate": 5.128958018758012e-06,
      "loss": 0.5928,
      "step": 840
    },
    {
      "epoch": 0.9258511640425245,
      "grad_norm": 0.3367156982421875,
      "learning_rate": 3.1133046533455947e-06,
      "loss": 0.6049,
      "step": 860
    },
    {
      "epoch": 0.9473825864621181,
      "grad_norm": 0.28718671202659607,
      "learning_rate": 1.592487189516212e-06,
      "loss": 0.6053,
      "step": 880
    },
    {
      "epoch": 0.9689140088817118,
      "grad_norm": 0.32894209027290344,
      "learning_rate": 5.742730000568908e-07,
      "loss": 0.6172,
      "step": 900
    },
    {
      "epoch": 0.9689140088817118,
      "eval_loss": 0.595556914806366,
      "eval_runtime": 367.6823,
      "eval_samples_per_second": 10.12,
      "eval_steps_per_second": 2.532,
      "step": 900
    },
    {
      "epoch": 0.9904454313013054,
      "grad_norm": 0.34396475553512573,
      "learning_rate": 6.386247842353754e-08,
      "loss": 0.6045,
      "step": 920
    }
  ],
  "logging_steps": 20,
  "max_steps": 929,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2554944340455096e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}