{
  "best_metric": 0.21792557956921524,
  "best_model_checkpoint": "/workspace/disk2/krishna/checkpoints/checkpoint-480",
  "epoch": 0.048,
  "eval_steps": 10,
  "global_step": 480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001,
      "grad_norm": 1.7856025695800781,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.682,
      "step": 10
    },
    {
      "epoch": 0.001,
      "eval_cos_sim": 0.30435964465141296,
      "eval_loss": 0.6965229106162732,
      "eval_runtime": 181.358,
      "eval_samples_per_second": 22.056,
      "eval_steps_per_second": 0.347,
      "step": 10
    },
    {
      "epoch": 0.002,
      "grad_norm": 1.632306694984436,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.6902,
      "step": 20
    },
    {
      "epoch": 0.002,
      "eval_cos_sim": 0.3060647249221802,
      "eval_loss": 0.6948180051063245,
      "eval_runtime": 158.9409,
      "eval_samples_per_second": 25.167,
      "eval_steps_per_second": 0.396,
      "step": 20
    },
    {
      "epoch": 0.003,
      "grad_norm": 1.4781064987182617,
      "learning_rate": 1.5e-06,
      "loss": 0.6879,
      "step": 30
    },
    {
      "epoch": 0.003,
      "eval_cos_sim": 0.30893880128860474,
      "eval_loss": 0.6919441828941053,
      "eval_runtime": 163.5656,
      "eval_samples_per_second": 24.455,
      "eval_steps_per_second": 0.385,
      "step": 30
    },
    {
      "epoch": 0.004,
      "grad_norm": 1.4996154308319092,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.6805,
      "step": 40
    },
    {
      "epoch": 0.004,
      "eval_cos_sim": 0.3130052089691162,
      "eval_loss": 0.6878782258247084,
      "eval_runtime": 160.8804,
      "eval_samples_per_second": 24.863,
      "eval_steps_per_second": 0.392,
      "step": 40
    },
    {
      "epoch": 0.005,
      "grad_norm": 1.7979084253311157,
      "learning_rate": 2.5e-06,
      "loss": 0.6795,
      "step": 50
    },
    {
      "epoch": 0.005,
      "eval_cos_sim": 0.3181459903717041,
      "eval_loss": 0.6827379918311781,
      "eval_runtime": 159.4021,
      "eval_samples_per_second": 25.094,
      "eval_steps_per_second": 0.395,
      "step": 50
    },
    {
      "epoch": 0.006,
      "grad_norm": 1.4049359560012817,
      "learning_rate": 3e-06,
      "loss": 0.6781,
      "step": 60
    },
    {
      "epoch": 0.006,
      "eval_cos_sim": 0.32444360852241516,
      "eval_loss": 0.6764405532096571,
      "eval_runtime": 159.9515,
      "eval_samples_per_second": 25.008,
      "eval_steps_per_second": 0.394,
      "step": 60
    },
    {
      "epoch": 0.007,
      "grad_norm": 1.7637789249420166,
      "learning_rate": 3.5000000000000004e-06,
      "loss": 0.6718,
      "step": 70
    },
    {
      "epoch": 0.007,
      "eval_cos_sim": 0.331728994846344,
      "eval_loss": 0.6691558842872327,
      "eval_runtime": 160.8921,
      "eval_samples_per_second": 24.861,
      "eval_steps_per_second": 0.392,
      "step": 70
    },
    {
      "epoch": 0.008,
      "grad_norm": 1.7560211420059204,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.6551,
      "step": 80
    },
    {
      "epoch": 0.008,
      "eval_cos_sim": 0.3400111794471741,
      "eval_loss": 0.6608743510459608,
      "eval_runtime": 172.4109,
      "eval_samples_per_second": 23.2,
      "eval_steps_per_second": 0.365,
      "step": 80
    },
    {
      "epoch": 0.009,
      "grad_norm": 1.7577630281448364,
      "learning_rate": 4.5e-06,
      "loss": 0.6566,
      "step": 90
    },
    {
      "epoch": 0.009,
      "eval_cos_sim": 0.34923624992370605,
      "eval_loss": 0.6516506343101209,
      "eval_runtime": 161.459,
      "eval_samples_per_second": 24.774,
      "eval_steps_per_second": 0.39,
      "step": 90
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.7723366022109985,
      "learning_rate": 5e-06,
      "loss": 0.6332,
      "step": 100
    },
    {
      "epoch": 0.01,
      "eval_cos_sim": 0.35936808586120605,
      "eval_loss": 0.6415200128768629,
      "eval_runtime": 158.9596,
      "eval_samples_per_second": 25.164,
      "eval_steps_per_second": 0.396,
      "step": 100
    },
    {
      "epoch": 0.011,
      "grad_norm": 1.7818301916122437,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.6297,
      "step": 110
    },
    {
      "epoch": 0.011,
      "eval_cos_sim": 0.3702617287635803,
      "eval_loss": 0.6306272282813734,
      "eval_runtime": 161.4739,
      "eval_samples_per_second": 24.772,
      "eval_steps_per_second": 0.39,
      "step": 110
    },
    {
      "epoch": 0.012,
      "grad_norm": 1.4825223684310913,
      "learning_rate": 6e-06,
      "loss": 0.6473,
      "step": 120
    },
    {
      "epoch": 0.012,
      "eval_cos_sim": 0.381907194852829,
      "eval_loss": 0.6189834599708265,
      "eval_runtime": 162.0142,
      "eval_samples_per_second": 24.689,
      "eval_steps_per_second": 0.389,
      "step": 120
    },
    {
      "epoch": 0.013,
      "grad_norm": 1.7424395084381104,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.611,
      "step": 130
    },
    {
      "epoch": 0.013,
      "eval_cos_sim": 0.3940889537334442,
      "eval_loss": 0.6068024029945082,
      "eval_runtime": 161.0467,
      "eval_samples_per_second": 24.838,
      "eval_steps_per_second": 0.391,
      "step": 130
    },
    {
      "epoch": 0.014,
      "grad_norm": 1.7474108934402466,
      "learning_rate": 7.000000000000001e-06,
      "loss": 0.5932,
      "step": 140
    },
    {
      "epoch": 0.014,
      "eval_cos_sim": 0.4073576033115387,
      "eval_loss": 0.5935340962623304,
      "eval_runtime": 161.0017,
      "eval_samples_per_second": 24.844,
      "eval_steps_per_second": 0.391,
      "step": 140
    },
    {
      "epoch": 0.015,
      "grad_norm": 1.6058062314987183,
      "learning_rate": 7.5e-06,
      "loss": 0.5824,
      "step": 150
    },
    {
      "epoch": 0.015,
      "eval_cos_sim": 0.42149588465690613,
      "eval_loss": 0.5793958764289564,
      "eval_runtime": 167.7429,
      "eval_samples_per_second": 23.846,
      "eval_steps_per_second": 0.376,
      "step": 150
    },
    {
      "epoch": 0.016,
      "grad_norm": 1.5788137912750244,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.5634,
      "step": 160
    },
    {
      "epoch": 0.016,
      "eval_cos_sim": 0.43618762493133545,
      "eval_loss": 0.564703805944795,
      "eval_runtime": 159.9946,
      "eval_samples_per_second": 25.001,
      "eval_steps_per_second": 0.394,
      "step": 160
    },
    {
      "epoch": 0.017,
      "grad_norm": 1.6113280057907104,
      "learning_rate": 8.500000000000002e-06,
      "loss": 0.5519,
      "step": 170
    },
    {
      "epoch": 0.017,
      "eval_cos_sim": 0.4508235454559326,
      "eval_loss": 0.5500685830329604,
      "eval_runtime": 157.8438,
      "eval_samples_per_second": 25.342,
      "eval_steps_per_second": 0.399,
      "step": 170
    },
    {
      "epoch": 0.018,
      "grad_norm": 1.6156154870986938,
      "learning_rate": 9e-06,
      "loss": 0.5407,
      "step": 180
    },
    {
      "epoch": 0.018,
      "eval_cos_sim": 0.46566617488861084,
      "eval_loss": 0.5352267394279188,
      "eval_runtime": 159.9444,
      "eval_samples_per_second": 25.009,
      "eval_steps_per_second": 0.394,
      "step": 180
    },
    {
      "epoch": 0.019,
      "grad_norm": 1.5958607196807861,
      "learning_rate": 9.5e-06,
      "loss": 0.5117,
      "step": 190
    },
    {
      "epoch": 0.019,
      "eval_cos_sim": 0.4806951582431793,
      "eval_loss": 0.5202000065063185,
      "eval_runtime": 157.8685,
      "eval_samples_per_second": 25.338,
      "eval_steps_per_second": 0.399,
      "step": 190
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.6012651920318604,
      "learning_rate": 1e-05,
      "loss": 0.5116,
      "step": 200
    },
    {
      "epoch": 0.02,
      "eval_cos_sim": 0.49558621644973755,
      "eval_loss": 0.5053114848350233,
      "eval_runtime": 155.395,
      "eval_samples_per_second": 25.741,
      "eval_steps_per_second": 0.405,
      "step": 200
    },
    {
      "epoch": 0.021,
      "grad_norm": 1.7965919971466064,
      "learning_rate": 1.05e-05,
      "loss": 0.4782,
      "step": 210
    },
    {
      "epoch": 0.021,
      "eval_cos_sim": 0.5108910202980042,
      "eval_loss": 0.49000780775296154,
      "eval_runtime": 159.6303,
      "eval_samples_per_second": 25.058,
      "eval_steps_per_second": 0.395,
      "step": 210
    },
    {
      "epoch": 0.022,
      "grad_norm": 1.791284441947937,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.4802,
      "step": 220
    },
    {
      "epoch": 0.022,
      "eval_cos_sim": 0.5271998643875122,
      "eval_loss": 0.4737012028907484,
      "eval_runtime": 169.963,
      "eval_samples_per_second": 23.535,
      "eval_steps_per_second": 0.371,
      "step": 220
    },
    {
      "epoch": 0.023,
      "grad_norm": 1.7923716306686401,
      "learning_rate": 1.1500000000000002e-05,
      "loss": 0.4513,
      "step": 230
    },
    {
      "epoch": 0.023,
      "eval_cos_sim": 0.5435076355934143,
      "eval_loss": 0.45739728691327036,
      "eval_runtime": 160.753,
      "eval_samples_per_second": 24.883,
      "eval_steps_per_second": 0.392,
      "step": 230
    },
    {
      "epoch": 0.024,
      "grad_norm": 1.7997322082519531,
      "learning_rate": 1.2e-05,
      "loss": 0.4288,
      "step": 240
    },
    {
      "epoch": 0.024,
      "eval_cos_sim": 0.5595658421516418,
      "eval_loss": 0.44134216930615366,
      "eval_runtime": 163.2069,
      "eval_samples_per_second": 24.509,
      "eval_steps_per_second": 0.386,
      "step": 240
    },
    {
      "epoch": 0.025,
      "grad_norm": 1.7595293521881104,
      "learning_rate": 1.25e-05,
      "loss": 0.4365,
      "step": 250
    },
    {
      "epoch": 0.025,
      "eval_cos_sim": 0.5746485590934753,
      "eval_loss": 0.4262613997672743,
      "eval_runtime": 160.8892,
      "eval_samples_per_second": 24.862,
      "eval_steps_per_second": 0.392,
      "step": 250
    },
    {
      "epoch": 0.026,
      "grad_norm": 1.614985704421997,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.4007,
      "step": 260
    },
    {
      "epoch": 0.026,
      "eval_cos_sim": 0.5887113809585571,
      "eval_loss": 0.4121987219070143,
      "eval_runtime": 157.6029,
      "eval_samples_per_second": 25.38,
      "eval_steps_per_second": 0.4,
      "step": 260
    },
    {
      "epoch": 0.027,
      "grad_norm": 1.6005358695983887,
      "learning_rate": 1.3500000000000001e-05,
      "loss": 0.4048,
      "step": 270
    },
    {
      "epoch": 0.027,
      "eval_cos_sim": 0.6018706560134888,
      "eval_loss": 0.39904179241406385,
      "eval_runtime": 163.2521,
      "eval_samples_per_second": 24.502,
      "eval_steps_per_second": 0.386,
      "step": 270
    },
    {
      "epoch": 0.028,
      "grad_norm": 1.7216198444366455,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.3793,
      "step": 280
    },
    {
      "epoch": 0.028,
      "eval_cos_sim": 0.6152481436729431,
      "eval_loss": 0.385668321153993,
      "eval_runtime": 163.9341,
      "eval_samples_per_second": 24.4,
      "eval_steps_per_second": 0.384,
      "step": 280
    },
    {
      "epoch": 0.029,
      "grad_norm": 1.7449027299880981,
      "learning_rate": 1.45e-05,
      "loss": 0.3614,
      "step": 290
    },
    {
      "epoch": 0.029,
      "eval_cos_sim": 0.6277655959129333,
      "eval_loss": 0.3731525359367079,
      "eval_runtime": 170.8176,
      "eval_samples_per_second": 23.417,
      "eval_steps_per_second": 0.369,
      "step": 290
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.5057954788208008,
      "learning_rate": 1.5e-05,
      "loss": 0.3557,
      "step": 300
    },
    {
      "epoch": 0.03,
      "eval_cos_sim": 0.6390805244445801,
      "eval_loss": 0.36183924534069956,
      "eval_runtime": 160.3353,
      "eval_samples_per_second": 24.948,
      "eval_steps_per_second": 0.393,
      "step": 300
    },
    {
      "epoch": 0.031,
      "grad_norm": 1.5245262384414673,
      "learning_rate": 1.55e-05,
      "loss": 0.3729,
      "step": 310
    },
    {
      "epoch": 0.031,
      "eval_cos_sim": 0.6503457427024841,
      "eval_loss": 0.3505764603828138,
      "eval_runtime": 161.447,
      "eval_samples_per_second": 24.776,
      "eval_steps_per_second": 0.39,
      "step": 310
    },
    {
      "epoch": 0.032,
      "grad_norm": 1.4508600234985352,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.3476,
      "step": 320
    },
    {
      "epoch": 0.032,
      "eval_cos_sim": 0.664078414440155,
      "eval_loss": 0.33684703972088753,
      "eval_runtime": 162.7312,
      "eval_samples_per_second": 24.58,
      "eval_steps_per_second": 0.387,
      "step": 320
    },
    {
      "epoch": 0.033,
      "grad_norm": 1.545847773551941,
      "learning_rate": 1.65e-05,
      "loss": 0.3335,
      "step": 330
    },
    {
      "epoch": 0.033,
      "eval_cos_sim": 0.6734265089035034,
      "eval_loss": 0.32750138094174325,
      "eval_runtime": 162.4883,
      "eval_samples_per_second": 24.617,
      "eval_steps_per_second": 0.388,
      "step": 330
    },
    {
      "epoch": 0.034,
      "grad_norm": 1.5199618339538574,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.3369,
      "step": 340
    },
    {
      "epoch": 0.034,
      "eval_cos_sim": 0.6810038685798645,
      "eval_loss": 0.31992570402371345,
      "eval_runtime": 161.4031,
      "eval_samples_per_second": 24.783,
      "eval_steps_per_second": 0.39,
      "step": 340
    },
    {
      "epoch": 0.035,
      "grad_norm": 1.2642650604248047,
      "learning_rate": 1.75e-05,
      "loss": 0.3396,
      "step": 350
    },
    {
      "epoch": 0.035,
      "eval_cos_sim": 0.6911267638206482,
      "eval_loss": 0.3098063874458021,
      "eval_runtime": 165.9265,
      "eval_samples_per_second": 24.107,
      "eval_steps_per_second": 0.38,
      "step": 350
    },
    {
      "epoch": 0.036,
      "grad_norm": 1.2891690731048584,
      "learning_rate": 1.8e-05,
      "loss": 0.3015,
      "step": 360
    },
    {
      "epoch": 0.036,
      "eval_cos_sim": 0.6987318992614746,
      "eval_loss": 0.3022033410285658,
      "eval_runtime": 173.6837,
      "eval_samples_per_second": 23.03,
      "eval_steps_per_second": 0.363,
      "step": 360
    },
    {
      "epoch": 0.037,
      "grad_norm": 1.2396866083145142,
      "learning_rate": 1.85e-05,
      "loss": 0.3077,
      "step": 370
    },
    {
      "epoch": 0.037,
      "eval_cos_sim": 0.7082959413528442,
      "eval_loss": 0.2926409912322706,
      "eval_runtime": 158.0401,
      "eval_samples_per_second": 25.31,
      "eval_steps_per_second": 0.399,
      "step": 370
    },
    {
      "epoch": 0.038,
      "grad_norm": 1.2055168151855469,
      "learning_rate": 1.9e-05,
      "loss": 0.2895,
      "step": 380
    },
    {
      "epoch": 0.038,
      "eval_cos_sim": 0.7168179750442505,
      "eval_loss": 0.284124677202577,
      "eval_runtime": 160.1742,
      "eval_samples_per_second": 24.973,
      "eval_steps_per_second": 0.393,
      "step": 380
    },
    {
      "epoch": 0.039,
      "grad_norm": 1.1191834211349487,
      "learning_rate": 1.9500000000000003e-05,
      "loss": 0.2944,
      "step": 390
    },
    {
      "epoch": 0.039,
      "eval_cos_sim": 0.7234470248222351,
      "eval_loss": 0.2775011363242811,
      "eval_runtime": 165.4196,
      "eval_samples_per_second": 24.181,
      "eval_steps_per_second": 0.381,
      "step": 390
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.097105860710144,
      "learning_rate": 2e-05,
      "loss": 0.2898,
      "step": 400
    },
    {
      "epoch": 0.04,
      "eval_cos_sim": 0.7314009666442871,
      "eval_loss": 0.2695479884360975,
      "eval_runtime": 159.5742,
      "eval_samples_per_second": 25.067,
      "eval_steps_per_second": 0.395,
      "step": 400
    },
    {
      "epoch": 0.041,
      "grad_norm": 1.040711760520935,
      "learning_rate": 2.05e-05,
      "loss": 0.2573,
      "step": 410
    },
    {
      "epoch": 0.041,
      "eval_cos_sim": 0.7389957904815674,
      "eval_loss": 0.26195128538357676,
      "eval_runtime": 160.3231,
      "eval_samples_per_second": 24.95,
      "eval_steps_per_second": 0.393,
      "step": 410
    },
    {
      "epoch": 0.042,
      "grad_norm": 1.0322086811065674,
      "learning_rate": 2.1e-05,
      "loss": 0.241,
      "step": 420
    },
    {
      "epoch": 0.042,
      "eval_cos_sim": 0.7471259236335754,
      "eval_loss": 0.25382448627697884,
      "eval_runtime": 161.5365,
      "eval_samples_per_second": 24.762,
      "eval_steps_per_second": 0.39,
      "step": 420
    },
    {
      "epoch": 0.043,
      "grad_norm": 0.9718736410140991,
      "learning_rate": 2.15e-05,
      "loss": 0.26,
      "step": 430
    },
    {
      "epoch": 0.043,
      "eval_cos_sim": 0.7534591555595398,
      "eval_loss": 0.24748690702664317,
      "eval_runtime": 169.2487,
      "eval_samples_per_second": 23.634,
      "eval_steps_per_second": 0.372,
      "step": 430
    },
    {
      "epoch": 0.044,
      "grad_norm": 0.8780993223190308,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.2255,
      "step": 440
    },
    {
      "epoch": 0.044,
      "eval_cos_sim": 0.7604101896286011,
      "eval_loss": 0.2405370173667616,
      "eval_runtime": 161.7354,
      "eval_samples_per_second": 24.732,
      "eval_steps_per_second": 0.39,
      "step": 440
    },
    {
      "epoch": 0.045,
      "grad_norm": 0.8193601369857788,
      "learning_rate": 2.25e-05,
      "loss": 0.2182,
      "step": 450
    },
    {
      "epoch": 0.045,
      "eval_cos_sim": 0.7673761248588562,
      "eval_loss": 0.2335676951621718,
      "eval_runtime": 159.2952,
      "eval_samples_per_second": 25.111,
      "eval_steps_per_second": 0.395,
      "step": 450
    },
    {
      "epoch": 0.046,
      "grad_norm": 0.8820355534553528,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 0.2153,
      "step": 460
    },
    {
      "epoch": 0.046,
      "eval_cos_sim": 0.7723870277404785,
      "eval_loss": 0.228554349443788,
      "eval_runtime": 161.8756,
      "eval_samples_per_second": 24.71,
      "eval_steps_per_second": 0.389,
      "step": 460
    },
    {
      "epoch": 0.047,
      "grad_norm": 0.9045171141624451,
      "learning_rate": 2.35e-05,
      "loss": 0.2245,
      "step": 470
    },
    {
      "epoch": 0.047,
      "eval_cos_sim": 0.7763780951499939,
      "eval_loss": 0.22456564476239152,
      "eval_runtime": 160.6855,
      "eval_samples_per_second": 24.893,
      "eval_steps_per_second": 0.392,
      "step": 470
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.7304402589797974,
      "learning_rate": 2.4e-05,
      "loss": 0.2069,
      "step": 480
    },
    {
      "epoch": 0.048,
      "eval_cos_sim": 0.7830195426940918,
      "eval_loss": 0.21792557956921524,
      "eval_runtime": 162.6536,
      "eval_samples_per_second": 24.592,
      "eval_steps_per_second": 0.387,
      "step": 480
    }
  ],
  "logging_steps": 10,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 10,
  "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": 160,
  "trial_name": null,
  "trial_params": null
}