small/checkpoints_v3/checkpoint-2000/trainer_state.json

{
  "best_global_step": 2000,
  "best_metric": 0.6828322410583496,
  "best_model_checkpoint": "cache/small/checkpoints_v3/checkpoint-2000",
  "epoch": 0.1025010250102501,
  "eval_steps": 2000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025625256252562525,
      "grad_norm": 1.1051346063613892,
      "learning_rate": 9.800000000000001e-06,
      "loss": 0.36488655090332034,
      "step": 50
    },
    {
      "epoch": 0.005125051250512505,
      "grad_norm": 1.0379502773284912,
      "learning_rate": 1.9800000000000004e-05,
      "loss": 0.35713844299316405,
      "step": 100
    },
    {
      "epoch": 0.007687576875768758,
      "grad_norm": 1.1215693950653076,
      "learning_rate": 2.98e-05,
      "loss": 0.3607748794555664,
      "step": 150
    },
    {
      "epoch": 0.01025010250102501,
      "grad_norm": 1.1428300142288208,
      "learning_rate": 3.9800000000000005e-05,
      "loss": 0.3552509307861328,
      "step": 200
    },
    {
      "epoch": 0.012812628126281263,
      "grad_norm": 1.1242265701293945,
      "learning_rate": 4.9800000000000004e-05,
      "loss": 0.3576408004760742,
      "step": 250
    },
    {
      "epoch": 0.015375153751537515,
      "grad_norm": 1.1097745895385742,
      "learning_rate": 5.9800000000000003e-05,
      "loss": 0.35824737548828123,
      "step": 300
    },
    {
      "epoch": 0.017937679376793768,
      "grad_norm": 1.0019261837005615,
      "learning_rate": 6.98e-05,
      "loss": 0.3588117599487305,
      "step": 350
    },
    {
      "epoch": 0.02050020500205002,
      "grad_norm": 0.9644513726234436,
      "learning_rate": 7.98e-05,
      "loss": 0.3657228469848633,
      "step": 400
    },
    {
      "epoch": 0.023062730627306273,
      "grad_norm": 1.1716028451919556,
      "learning_rate": 8.98e-05,
      "loss": 0.37080657958984375,
      "step": 450
    },
    {
      "epoch": 0.025625256252562525,
      "grad_norm": 1.018113613128662,
      "learning_rate": 9.98e-05,
      "loss": 0.36805721282958986,
      "step": 500
    },
    {
      "epoch": 0.028187781877818778,
      "grad_norm": 1.1356464624404907,
      "learning_rate": 9.99495157634453e-05,
      "loss": 0.37148025512695315,
      "step": 550
    },
    {
      "epoch": 0.03075030750307503,
      "grad_norm": 1.2209874391555786,
      "learning_rate": 9.989800123634865e-05,
      "loss": 0.3799387359619141,
      "step": 600
    },
    {
      "epoch": 0.03331283312833128,
      "grad_norm": 1.0993411540985107,
      "learning_rate": 9.984648670925202e-05,
      "loss": 0.38206844329833983,
      "step": 650
    },
    {
      "epoch": 0.035875358753587536,
      "grad_norm": 1.150312066078186,
      "learning_rate": 9.979497218215538e-05,
      "loss": 0.3814669036865234,
      "step": 700
    },
    {
      "epoch": 0.03843788437884379,
      "grad_norm": 1.1544060707092285,
      "learning_rate": 9.974345765505873e-05,
      "loss": 0.3826092529296875,
      "step": 750
    },
    {
      "epoch": 0.04100041000410004,
      "grad_norm": 1.075146198272705,
      "learning_rate": 9.96919431279621e-05,
      "loss": 0.377677001953125,
      "step": 800
    },
    {
      "epoch": 0.04356293562935629,
      "grad_norm": 1.070531964302063,
      "learning_rate": 9.964042860086545e-05,
      "loss": 0.38505050659179685,
      "step": 850
    },
    {
      "epoch": 0.046125461254612546,
      "grad_norm": 1.1864845752716064,
      "learning_rate": 9.958891407376881e-05,
      "loss": 0.3866250991821289,
      "step": 900
    },
    {
      "epoch": 0.0486879868798688,
      "grad_norm": 1.229409098625183,
      "learning_rate": 9.953739954667216e-05,
      "loss": 0.390551872253418,
      "step": 950
    },
    {
      "epoch": 0.05125051250512505,
      "grad_norm": 1.10013747215271,
      "learning_rate": 9.948588501957552e-05,
      "loss": 0.3865232086181641,
      "step": 1000
    },
    {
      "epoch": 0.053813038130381304,
      "grad_norm": 1.2265090942382812,
      "learning_rate": 9.943437049247888e-05,
      "loss": 0.38285724639892577,
      "step": 1050
    },
    {
      "epoch": 0.056375563755637556,
      "grad_norm": 1.1173005104064941,
      "learning_rate": 9.938285596538224e-05,
      "loss": 0.3867513275146484,
      "step": 1100
    },
    {
      "epoch": 0.05893808938089381,
      "grad_norm": 1.1064320802688599,
      "learning_rate": 9.933134143828559e-05,
      "loss": 0.3864853286743164,
      "step": 1150
    },
    {
      "epoch": 0.06150061500615006,
      "grad_norm": 0.9700886607170105,
      "learning_rate": 9.927982691118896e-05,
      "loss": 0.3844454574584961,
      "step": 1200
    },
    {
      "epoch": 0.06406314063140632,
      "grad_norm": 1.0213518142700195,
      "learning_rate": 9.922831238409231e-05,
      "loss": 0.39336017608642576,
      "step": 1250
    },
    {
      "epoch": 0.06662566625666257,
      "grad_norm": 1.0725339651107788,
      "learning_rate": 9.917679785699567e-05,
      "loss": 0.38393234252929687,
      "step": 1300
    },
    {
      "epoch": 0.06918819188191883,
      "grad_norm": 1.1702908277511597,
      "learning_rate": 9.912528332989904e-05,
      "loss": 0.3882109069824219,
      "step": 1350
    },
    {
      "epoch": 0.07175071750717507,
      "grad_norm": 1.082552194595337,
      "learning_rate": 9.907376880280239e-05,
      "loss": 0.3867752838134766,
      "step": 1400
    },
    {
      "epoch": 0.07431324313243133,
      "grad_norm": 1.128678798675537,
      "learning_rate": 9.902225427570574e-05,
      "loss": 0.3881544494628906,
      "step": 1450
    },
    {
      "epoch": 0.07687576875768758,
      "grad_norm": 1.0918948650360107,
      "learning_rate": 9.897073974860911e-05,
      "loss": 0.3869658660888672,
      "step": 1500
    },
    {
      "epoch": 0.07943829438294384,
      "grad_norm": 1.0815236568450928,
      "learning_rate": 9.891922522151247e-05,
      "loss": 0.39020267486572263,
      "step": 1550
    },
    {
      "epoch": 0.08200082000820008,
      "grad_norm": 1.0975167751312256,
      "learning_rate": 9.886771069441582e-05,
      "loss": 0.3906606292724609,
      "step": 1600
    },
    {
      "epoch": 0.08456334563345634,
      "grad_norm": 1.116685152053833,
      "learning_rate": 9.881619616731919e-05,
      "loss": 0.3883519744873047,
      "step": 1650
    },
    {
      "epoch": 0.08712587125871259,
      "grad_norm": 1.0836381912231445,
      "learning_rate": 9.876468164022254e-05,
      "loss": 0.38965553283691406,
      "step": 1700
    },
    {
      "epoch": 0.08968839688396885,
      "grad_norm": 1.043843388557434,
      "learning_rate": 9.871316711312591e-05,
      "loss": 0.3916368103027344,
      "step": 1750
    },
    {
      "epoch": 0.09225092250922509,
      "grad_norm": 1.1080198287963867,
      "learning_rate": 9.866165258602926e-05,
      "loss": 0.3878855895996094,
      "step": 1800
    },
    {
      "epoch": 0.09481344813448135,
      "grad_norm": 1.1533907651901245,
      "learning_rate": 9.861013805893262e-05,
      "loss": 0.38703533172607424,
      "step": 1850
    },
    {
      "epoch": 0.0973759737597376,
      "grad_norm": 1.0471681356430054,
      "learning_rate": 9.855862353183599e-05,
      "loss": 0.3946280670166016,
      "step": 1900
    },
    {
      "epoch": 0.09993849938499386,
      "grad_norm": 1.069541335105896,
      "learning_rate": 9.850710900473934e-05,
      "loss": 0.38905384063720705,
      "step": 1950
    },
    {
      "epoch": 0.1025010250102501,
      "grad_norm": 1.1811953783035278,
      "learning_rate": 9.84555944776427e-05,
      "loss": 0.3862751388549805,
      "step": 2000
    },
    {
      "epoch": 0.1025010250102501,
      "eval_loss": 0.7286977767944336,
      "eval_runtime": 2.0545,
      "eval_samples_per_second": 4867.301,
      "eval_steps_per_second": 4.867,
      "eval_teacher_cosine_loss": 0.3171677589416504,
      "eval_teacher_cosine_mean": 0.6828322410583496,
      "eval_teacher_cosine_p10": 0.5881704092025757,
      "eval_teacher_cosine_p50": 0.6930590867996216,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 97560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 512,
  "trial_name": null,
  "trial_params": null
}