sheepy928's picture
Training in progress, step 2500, checkpoint
967b7ef verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.208313194780087,
"eval_steps": 500,
"global_step": 2500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": 75.8236083984375,
"learning_rate": 4.0000000000000003e-07,
"loss": 3.7148,
"step": 10
},
{
"epoch": 0.01,
"grad_norm": 52.23726272583008,
"learning_rate": 9e-07,
"loss": 3.2721,
"step": 20
},
{
"epoch": 0.01,
"grad_norm": 35.391998291015625,
"learning_rate": 1.4000000000000001e-06,
"loss": 2.4262,
"step": 30
},
{
"epoch": 0.02,
"grad_norm": 15.225934028625488,
"learning_rate": 1.9e-06,
"loss": 1.8828,
"step": 40
},
{
"epoch": 0.02,
"grad_norm": 13.488654136657715,
"learning_rate": 2.4000000000000003e-06,
"loss": 1.5398,
"step": 50
},
{
"epoch": 0.03,
"grad_norm": 8.773133277893066,
"learning_rate": 2.8500000000000002e-06,
"loss": 1.2772,
"step": 60
},
{
"epoch": 0.03,
"grad_norm": 6.736795902252197,
"learning_rate": 3.3500000000000005e-06,
"loss": 1.1159,
"step": 70
},
{
"epoch": 0.04,
"grad_norm": 2.9463019371032715,
"learning_rate": 3.85e-06,
"loss": 0.9883,
"step": 80
},
{
"epoch": 0.04,
"grad_norm": 1.3851969242095947,
"learning_rate": 4.35e-06,
"loss": 0.902,
"step": 90
},
{
"epoch": 0.05,
"grad_norm": 1.4642181396484375,
"learning_rate": 4.85e-06,
"loss": 0.8701,
"step": 100
},
{
"epoch": 0.05,
"grad_norm": 0.6514687538146973,
"learning_rate": 5.3500000000000004e-06,
"loss": 0.8548,
"step": 110
},
{
"epoch": 0.06,
"grad_norm": 0.4448617994785309,
"learning_rate": 5.850000000000001e-06,
"loss": 0.8038,
"step": 120
},
{
"epoch": 0.06,
"grad_norm": 0.5425733327865601,
"learning_rate": 6.35e-06,
"loss": 0.825,
"step": 130
},
{
"epoch": 0.07,
"grad_norm": 0.47352057695388794,
"learning_rate": 6.8500000000000005e-06,
"loss": 0.812,
"step": 140
},
{
"epoch": 0.07,
"grad_norm": 0.40401744842529297,
"learning_rate": 7.35e-06,
"loss": 0.7912,
"step": 150
},
{
"epoch": 0.08,
"grad_norm": 0.3112718164920807,
"learning_rate": 7.850000000000001e-06,
"loss": 0.8057,
"step": 160
},
{
"epoch": 0.08,
"grad_norm": 0.45351892709732056,
"learning_rate": 8.350000000000001e-06,
"loss": 0.7912,
"step": 170
},
{
"epoch": 0.09,
"grad_norm": 0.38080641627311707,
"learning_rate": 8.85e-06,
"loss": 0.7956,
"step": 180
},
{
"epoch": 0.09,
"grad_norm": 0.22949431836605072,
"learning_rate": 9.35e-06,
"loss": 0.7785,
"step": 190
},
{
"epoch": 0.1,
"grad_norm": 0.8331744074821472,
"learning_rate": 9.85e-06,
"loss": 0.7769,
"step": 200
},
{
"epoch": 0.1,
"grad_norm": 0.30213871598243713,
"learning_rate": 1.035e-05,
"loss": 0.7797,
"step": 210
},
{
"epoch": 0.11,
"grad_norm": 0.32015591859817505,
"learning_rate": 1.0850000000000001e-05,
"loss": 0.7941,
"step": 220
},
{
"epoch": 0.11,
"grad_norm": 0.22447015345096588,
"learning_rate": 1.1350000000000001e-05,
"loss": 0.8014,
"step": 230
},
{
"epoch": 0.12,
"grad_norm": 0.2546006143093109,
"learning_rate": 1.185e-05,
"loss": 0.771,
"step": 240
},
{
"epoch": 0.12,
"grad_norm": 0.23833389580249786,
"learning_rate": 1.235e-05,
"loss": 0.7956,
"step": 250
},
{
"epoch": 0.13,
"grad_norm": 0.1831669956445694,
"learning_rate": 1.285e-05,
"loss": 0.7833,
"step": 260
},
{
"epoch": 0.13,
"grad_norm": 0.19860319793224335,
"learning_rate": 1.3350000000000001e-05,
"loss": 0.7956,
"step": 270
},
{
"epoch": 0.14,
"grad_norm": 0.21220380067825317,
"learning_rate": 1.3850000000000001e-05,
"loss": 0.7942,
"step": 280
},
{
"epoch": 0.14,
"grad_norm": 0.26025038957595825,
"learning_rate": 1.435e-05,
"loss": 0.7877,
"step": 290
},
{
"epoch": 0.14,
"grad_norm": 0.2029470056295395,
"learning_rate": 1.485e-05,
"loss": 0.7637,
"step": 300
},
{
"epoch": 0.15,
"grad_norm": 0.21639183163642883,
"learning_rate": 1.535e-05,
"loss": 0.7885,
"step": 310
},
{
"epoch": 0.15,
"grad_norm": 0.23670387268066406,
"learning_rate": 1.5850000000000002e-05,
"loss": 0.7558,
"step": 320
},
{
"epoch": 0.16,
"grad_norm": 0.2855335772037506,
"learning_rate": 1.635e-05,
"loss": 0.7854,
"step": 330
},
{
"epoch": 0.16,
"grad_norm": 0.21515759825706482,
"learning_rate": 1.6850000000000003e-05,
"loss": 0.7577,
"step": 340
},
{
"epoch": 0.17,
"grad_norm": 0.18981383740901947,
"learning_rate": 1.7349999999999998e-05,
"loss": 0.7802,
"step": 350
},
{
"epoch": 0.17,
"grad_norm": 0.2674611508846283,
"learning_rate": 1.785e-05,
"loss": 0.7809,
"step": 360
},
{
"epoch": 0.18,
"grad_norm": 0.24821774661540985,
"learning_rate": 1.8350000000000002e-05,
"loss": 0.7965,
"step": 370
},
{
"epoch": 0.18,
"grad_norm": 0.3644929528236389,
"learning_rate": 1.885e-05,
"loss": 0.7957,
"step": 380
},
{
"epoch": 0.19,
"grad_norm": 0.3568972647190094,
"learning_rate": 1.9350000000000003e-05,
"loss": 0.7882,
"step": 390
},
{
"epoch": 0.19,
"grad_norm": 0.3236319422721863,
"learning_rate": 1.985e-05,
"loss": 0.7841,
"step": 400
},
{
"epoch": 0.2,
"grad_norm": 0.24553723633289337,
"learning_rate": 2.035e-05,
"loss": 0.7824,
"step": 410
},
{
"epoch": 0.2,
"grad_norm": 0.1855400949716568,
"learning_rate": 2.085e-05,
"loss": 0.7645,
"step": 420
},
{
"epoch": 0.21,
"grad_norm": 0.26212435960769653,
"learning_rate": 2.135e-05,
"loss": 0.7824,
"step": 430
},
{
"epoch": 0.21,
"grad_norm": 0.3167509138584137,
"learning_rate": 2.1850000000000003e-05,
"loss": 0.7908,
"step": 440
},
{
"epoch": 0.22,
"grad_norm": 0.21845023334026337,
"learning_rate": 2.235e-05,
"loss": 0.7888,
"step": 450
},
{
"epoch": 0.22,
"grad_norm": 0.30318892002105713,
"learning_rate": 2.2850000000000003e-05,
"loss": 0.7682,
"step": 460
},
{
"epoch": 0.23,
"grad_norm": 0.28828638792037964,
"learning_rate": 2.3350000000000002e-05,
"loss": 0.7588,
"step": 470
},
{
"epoch": 0.23,
"grad_norm": 0.2422240674495697,
"learning_rate": 2.385e-05,
"loss": 0.7787,
"step": 480
},
{
"epoch": 0.24,
"grad_norm": 0.3579326272010803,
"learning_rate": 2.435e-05,
"loss": 0.7932,
"step": 490
},
{
"epoch": 0.24,
"grad_norm": 0.3549540042877197,
"learning_rate": 2.485e-05,
"loss": 0.7634,
"step": 500
},
{
"epoch": 0.24,
"eval_loss": 0.7744565606117249,
"eval_runtime": 43.9626,
"eval_samples_per_second": 45.493,
"eval_steps_per_second": 0.364,
"step": 500
},
{
"epoch": 0.25,
"grad_norm": 0.2931281328201294,
"learning_rate": 2.5350000000000003e-05,
"loss": 0.7763,
"step": 510
},
{
"epoch": 0.25,
"grad_norm": 0.21228083968162537,
"learning_rate": 2.585e-05,
"loss": 0.7748,
"step": 520
},
{
"epoch": 0.26,
"grad_norm": 0.26342302560806274,
"learning_rate": 2.6350000000000004e-05,
"loss": 0.789,
"step": 530
},
{
"epoch": 0.26,
"grad_norm": 0.2840302288532257,
"learning_rate": 2.6850000000000002e-05,
"loss": 0.768,
"step": 540
},
{
"epoch": 0.27,
"grad_norm": 0.273771196603775,
"learning_rate": 2.7350000000000004e-05,
"loss": 0.7707,
"step": 550
},
{
"epoch": 0.27,
"grad_norm": 1.6022837162017822,
"learning_rate": 2.7850000000000003e-05,
"loss": 0.7705,
"step": 560
},
{
"epoch": 0.28,
"grad_norm": 0.3082919716835022,
"learning_rate": 2.8349999999999998e-05,
"loss": 0.7849,
"step": 570
},
{
"epoch": 0.28,
"grad_norm": 0.25341132283210754,
"learning_rate": 2.885e-05,
"loss": 0.7658,
"step": 580
},
{
"epoch": 0.29,
"grad_norm": 0.2250708043575287,
"learning_rate": 2.935e-05,
"loss": 0.756,
"step": 590
},
{
"epoch": 0.29,
"grad_norm": 0.23522531986236572,
"learning_rate": 2.985e-05,
"loss": 0.7692,
"step": 600
},
{
"epoch": 0.29,
"grad_norm": 0.2448827624320984,
"learning_rate": 3.035e-05,
"loss": 0.7635,
"step": 610
},
{
"epoch": 0.3,
"grad_norm": 0.19717586040496826,
"learning_rate": 3.0850000000000004e-05,
"loss": 0.7548,
"step": 620
},
{
"epoch": 0.3,
"grad_norm": 0.24299098551273346,
"learning_rate": 3.135e-05,
"loss": 0.7673,
"step": 630
},
{
"epoch": 0.31,
"grad_norm": 0.2555953562259674,
"learning_rate": 3.185e-05,
"loss": 0.7528,
"step": 640
},
{
"epoch": 0.31,
"grad_norm": 0.2102127969264984,
"learning_rate": 3.235e-05,
"loss": 0.7678,
"step": 650
},
{
"epoch": 0.32,
"grad_norm": 0.21767646074295044,
"learning_rate": 3.2850000000000006e-05,
"loss": 0.7564,
"step": 660
},
{
"epoch": 0.32,
"grad_norm": 0.3582068085670471,
"learning_rate": 3.3350000000000004e-05,
"loss": 0.7645,
"step": 670
},
{
"epoch": 0.33,
"grad_norm": 0.27804431319236755,
"learning_rate": 3.385e-05,
"loss": 0.7645,
"step": 680
},
{
"epoch": 0.33,
"grad_norm": 0.2464047521352768,
"learning_rate": 3.435e-05,
"loss": 0.7325,
"step": 690
},
{
"epoch": 0.34,
"grad_norm": 0.22323842346668243,
"learning_rate": 3.485e-05,
"loss": 0.7579,
"step": 700
},
{
"epoch": 0.34,
"grad_norm": 0.28329703211784363,
"learning_rate": 3.535e-05,
"loss": 0.7399,
"step": 710
},
{
"epoch": 0.35,
"grad_norm": 0.2471865713596344,
"learning_rate": 3.585e-05,
"loss": 0.765,
"step": 720
},
{
"epoch": 0.35,
"grad_norm": 0.314908504486084,
"learning_rate": 3.635e-05,
"loss": 0.7529,
"step": 730
},
{
"epoch": 0.36,
"grad_norm": 0.21466995775699615,
"learning_rate": 3.685e-05,
"loss": 0.7365,
"step": 740
},
{
"epoch": 0.36,
"grad_norm": 0.2545580565929413,
"learning_rate": 3.735e-05,
"loss": 0.7458,
"step": 750
},
{
"epoch": 0.37,
"grad_norm": 0.2651059329509735,
"learning_rate": 3.7850000000000005e-05,
"loss": 0.7578,
"step": 760
},
{
"epoch": 0.37,
"grad_norm": 0.2408379763364792,
"learning_rate": 3.8350000000000004e-05,
"loss": 0.7283,
"step": 770
},
{
"epoch": 0.38,
"grad_norm": 0.2575875520706177,
"learning_rate": 3.885e-05,
"loss": 0.7367,
"step": 780
},
{
"epoch": 0.38,
"grad_norm": 0.28182050585746765,
"learning_rate": 3.935e-05,
"loss": 0.7784,
"step": 790
},
{
"epoch": 0.39,
"grad_norm": 0.20579397678375244,
"learning_rate": 3.9850000000000006e-05,
"loss": 0.7533,
"step": 800
},
{
"epoch": 0.39,
"grad_norm": 0.1906532198190689,
"learning_rate": 4.0350000000000005e-05,
"loss": 0.7544,
"step": 810
},
{
"epoch": 0.4,
"grad_norm": 0.18311142921447754,
"learning_rate": 4.085e-05,
"loss": 0.75,
"step": 820
},
{
"epoch": 0.4,
"grad_norm": 0.1910839080810547,
"learning_rate": 4.135e-05,
"loss": 0.7538,
"step": 830
},
{
"epoch": 0.41,
"grad_norm": 0.2178005427122116,
"learning_rate": 4.185e-05,
"loss": 0.7456,
"step": 840
},
{
"epoch": 0.41,
"grad_norm": 0.27642378211021423,
"learning_rate": 4.235e-05,
"loss": 0.744,
"step": 850
},
{
"epoch": 0.42,
"grad_norm": 0.26685312390327454,
"learning_rate": 4.285e-05,
"loss": 0.7611,
"step": 860
},
{
"epoch": 0.42,
"grad_norm": 0.2021845579147339,
"learning_rate": 4.335e-05,
"loss": 0.758,
"step": 870
},
{
"epoch": 0.43,
"grad_norm": 0.20516808331012726,
"learning_rate": 4.385e-05,
"loss": 0.7404,
"step": 880
},
{
"epoch": 0.43,
"grad_norm": 0.20619668066501617,
"learning_rate": 4.435e-05,
"loss": 0.7379,
"step": 890
},
{
"epoch": 0.43,
"grad_norm": 0.22439967095851898,
"learning_rate": 4.4850000000000006e-05,
"loss": 0.7495,
"step": 900
},
{
"epoch": 0.44,
"grad_norm": 0.2192295491695404,
"learning_rate": 4.5350000000000005e-05,
"loss": 0.7377,
"step": 910
},
{
"epoch": 0.44,
"grad_norm": 0.26503369212150574,
"learning_rate": 4.585e-05,
"loss": 0.7481,
"step": 920
},
{
"epoch": 0.45,
"grad_norm": 0.2249353975057602,
"learning_rate": 4.635e-05,
"loss": 0.742,
"step": 930
},
{
"epoch": 0.45,
"grad_norm": 0.21413160860538483,
"learning_rate": 4.685000000000001e-05,
"loss": 0.7485,
"step": 940
},
{
"epoch": 0.46,
"grad_norm": 0.20971478521823883,
"learning_rate": 4.735e-05,
"loss": 0.7339,
"step": 950
},
{
"epoch": 0.46,
"grad_norm": 0.20295798778533936,
"learning_rate": 4.785e-05,
"loss": 0.722,
"step": 960
},
{
"epoch": 0.47,
"grad_norm": 0.1957596242427826,
"learning_rate": 4.835e-05,
"loss": 0.7538,
"step": 970
},
{
"epoch": 0.47,
"grad_norm": 0.18014560639858246,
"learning_rate": 4.885e-05,
"loss": 0.7251,
"step": 980
},
{
"epoch": 0.48,
"grad_norm": 0.19334861636161804,
"learning_rate": 4.935e-05,
"loss": 0.7444,
"step": 990
},
{
"epoch": 0.48,
"grad_norm": 0.1753440499305725,
"learning_rate": 4.9850000000000006e-05,
"loss": 0.7376,
"step": 1000
},
{
"epoch": 0.48,
"eval_loss": 0.73988938331604,
"eval_runtime": 42.9535,
"eval_samples_per_second": 46.562,
"eval_steps_per_second": 0.372,
"step": 1000
},
{
"epoch": 0.49,
"grad_norm": 0.24455028772354126,
"learning_rate": 4.99999307775404e-05,
"loss": 0.7404,
"step": 1010
},
{
"epoch": 0.49,
"grad_norm": 0.2149570733308792,
"learning_rate": 4.999959172968145e-05,
"loss": 0.7342,
"step": 1020
},
{
"epoch": 0.5,
"grad_norm": 0.1931283175945282,
"learning_rate": 4.999897014592085e-05,
"loss": 0.735,
"step": 1030
},
{
"epoch": 0.5,
"grad_norm": 0.2331906110048294,
"learning_rate": 4.999806603328352e-05,
"loss": 0.7145,
"step": 1040
},
{
"epoch": 0.51,
"grad_norm": 0.21177324652671814,
"learning_rate": 4.999687940198738e-05,
"loss": 0.7308,
"step": 1050
},
{
"epoch": 0.51,
"grad_norm": 0.2026621401309967,
"learning_rate": 4.9995410265443286e-05,
"loss": 0.7445,
"step": 1060
},
{
"epoch": 0.52,
"grad_norm": 0.19811251759529114,
"learning_rate": 4.999365864025487e-05,
"loss": 0.738,
"step": 1070
},
{
"epoch": 0.52,
"grad_norm": 0.1981566846370697,
"learning_rate": 4.9991624546218334e-05,
"loss": 0.7286,
"step": 1080
},
{
"epoch": 0.53,
"grad_norm": 0.17122291028499603,
"learning_rate": 4.9989308006322235e-05,
"loss": 0.737,
"step": 1090
},
{
"epoch": 0.53,
"grad_norm": 0.2182038575410843,
"learning_rate": 4.9986709046747225e-05,
"loss": 0.7354,
"step": 1100
},
{
"epoch": 0.54,
"grad_norm": 0.1823730617761612,
"learning_rate": 4.9983827696865764e-05,
"loss": 0.725,
"step": 1110
},
{
"epoch": 0.54,
"grad_norm": 0.2195768803358078,
"learning_rate": 4.998066398924177e-05,
"loss": 0.7173,
"step": 1120
},
{
"epoch": 0.55,
"grad_norm": 0.18349726498126984,
"learning_rate": 4.997721795963028e-05,
"loss": 0.7159,
"step": 1130
},
{
"epoch": 0.55,
"grad_norm": 0.2440386265516281,
"learning_rate": 4.997348964697699e-05,
"loss": 0.7171,
"step": 1140
},
{
"epoch": 0.56,
"grad_norm": 0.20371629297733307,
"learning_rate": 4.9969479093417894e-05,
"loss": 0.7179,
"step": 1150
},
{
"epoch": 0.56,
"grad_norm": 0.2982828617095947,
"learning_rate": 4.9965186344278746e-05,
"loss": 0.7222,
"step": 1160
},
{
"epoch": 0.57,
"grad_norm": 0.2097097784280777,
"learning_rate": 4.9960611448074555e-05,
"loss": 0.7174,
"step": 1170
},
{
"epoch": 0.57,
"grad_norm": 0.2264368236064911,
"learning_rate": 4.995575445650907e-05,
"loss": 0.7079,
"step": 1180
},
{
"epoch": 0.58,
"grad_norm": 0.24737605452537537,
"learning_rate": 4.995061542447418e-05,
"loss": 0.7272,
"step": 1190
},
{
"epoch": 0.58,
"grad_norm": 0.19530241191387177,
"learning_rate": 4.9945194410049254e-05,
"loss": 0.7216,
"step": 1200
},
{
"epoch": 0.58,
"grad_norm": 0.2791181802749634,
"learning_rate": 4.993949147450055e-05,
"loss": 0.712,
"step": 1210
},
{
"epoch": 0.59,
"grad_norm": 0.2095661610364914,
"learning_rate": 4.993350668228049e-05,
"loss": 0.7067,
"step": 1220
},
{
"epoch": 0.59,
"grad_norm": 0.17129307985305786,
"learning_rate": 4.9927240101026915e-05,
"loss": 0.7087,
"step": 1230
},
{
"epoch": 0.6,
"grad_norm": 0.2154540866613388,
"learning_rate": 4.992069180156235e-05,
"loss": 0.7411,
"step": 1240
},
{
"epoch": 0.6,
"grad_norm": 0.22642803192138672,
"learning_rate": 4.991386185789319e-05,
"loss": 0.7266,
"step": 1250
},
{
"epoch": 0.61,
"grad_norm": 0.19676081836223602,
"learning_rate": 4.9906750347208866e-05,
"loss": 0.7012,
"step": 1260
},
{
"epoch": 0.61,
"grad_norm": 0.2095811814069748,
"learning_rate": 4.989935734988098e-05,
"loss": 0.7162,
"step": 1270
},
{
"epoch": 0.62,
"grad_norm": 0.21455398201942444,
"learning_rate": 4.9891682949462385e-05,
"loss": 0.7177,
"step": 1280
},
{
"epoch": 0.62,
"grad_norm": 0.20205985009670258,
"learning_rate": 4.988372723268623e-05,
"loss": 0.718,
"step": 1290
},
{
"epoch": 0.63,
"grad_norm": 0.18780440092086792,
"learning_rate": 4.987549028946502e-05,
"loss": 0.7103,
"step": 1300
},
{
"epoch": 0.63,
"grad_norm": 0.20362792909145355,
"learning_rate": 4.986697221288956e-05,
"loss": 0.7304,
"step": 1310
},
{
"epoch": 0.64,
"grad_norm": 0.2820493280887604,
"learning_rate": 4.985817309922792e-05,
"loss": 0.7182,
"step": 1320
},
{
"epoch": 0.64,
"grad_norm": 0.20787598192691803,
"learning_rate": 4.984909304792435e-05,
"loss": 0.7142,
"step": 1330
},
{
"epoch": 0.65,
"grad_norm": 0.21624509990215302,
"learning_rate": 4.983973216159813e-05,
"loss": 0.7208,
"step": 1340
},
{
"epoch": 0.65,
"grad_norm": 0.19343602657318115,
"learning_rate": 4.983009054604246e-05,
"loss": 0.7097,
"step": 1350
},
{
"epoch": 0.66,
"grad_norm": 0.254402756690979,
"learning_rate": 4.9820168310223215e-05,
"loss": 0.7193,
"step": 1360
},
{
"epoch": 0.66,
"grad_norm": 0.1940561830997467,
"learning_rate": 4.980996556627774e-05,
"loss": 0.6988,
"step": 1370
},
{
"epoch": 0.67,
"grad_norm": 0.22593462467193604,
"learning_rate": 4.9799482429513576e-05,
"loss": 0.7054,
"step": 1380
},
{
"epoch": 0.67,
"grad_norm": 0.22020135819911957,
"learning_rate": 4.978871901840716e-05,
"loss": 0.7221,
"step": 1390
},
{
"epoch": 0.68,
"grad_norm": 0.22807644307613373,
"learning_rate": 4.977767545460248e-05,
"loss": 0.7213,
"step": 1400
},
{
"epoch": 0.68,
"grad_norm": 0.1925562173128128,
"learning_rate": 4.9766351862909703e-05,
"loss": 0.7127,
"step": 1410
},
{
"epoch": 0.69,
"grad_norm": 0.20845136046409607,
"learning_rate": 4.9754748371303775e-05,
"loss": 0.7207,
"step": 1420
},
{
"epoch": 0.69,
"grad_norm": 0.18583528697490692,
"learning_rate": 4.974286511092294e-05,
"loss": 0.7098,
"step": 1430
},
{
"epoch": 0.7,
"grad_norm": 0.2019319236278534,
"learning_rate": 4.973070221606732e-05,
"loss": 0.7029,
"step": 1440
},
{
"epoch": 0.7,
"grad_norm": 0.1966458261013031,
"learning_rate": 4.971825982419731e-05,
"loss": 0.7241,
"step": 1450
},
{
"epoch": 0.71,
"grad_norm": 0.20924623310565948,
"learning_rate": 4.970553807593214e-05,
"loss": 0.7126,
"step": 1460
},
{
"epoch": 0.71,
"grad_norm": 0.20655803382396698,
"learning_rate": 4.969253711504814e-05,
"loss": 0.6903,
"step": 1470
},
{
"epoch": 0.72,
"grad_norm": 0.2179209142923355,
"learning_rate": 4.967925708847727e-05,
"loss": 0.683,
"step": 1480
},
{
"epoch": 0.72,
"grad_norm": 0.2198410928249359,
"learning_rate": 4.966569814630534e-05,
"loss": 0.7087,
"step": 1490
},
{
"epoch": 0.72,
"grad_norm": 0.17970815300941467,
"learning_rate": 4.9651860441770374e-05,
"loss": 0.7006,
"step": 1500
},
{
"epoch": 0.72,
"eval_loss": 0.7137542366981506,
"eval_runtime": 43.0052,
"eval_samples_per_second": 46.506,
"eval_steps_per_second": 0.372,
"step": 1500
},
{
"epoch": 0.73,
"grad_norm": 0.1931023746728897,
"learning_rate": 4.963774413126086e-05,
"loss": 0.7245,
"step": 1510
},
{
"epoch": 0.73,
"grad_norm": 0.21812495589256287,
"learning_rate": 4.9623349374313973e-05,
"loss": 0.7119,
"step": 1520
},
{
"epoch": 0.74,
"grad_norm": 0.20784322917461395,
"learning_rate": 4.960867633361381e-05,
"loss": 0.704,
"step": 1530
},
{
"epoch": 0.74,
"grad_norm": 0.18697650730609894,
"learning_rate": 4.9593725174989507e-05,
"loss": 0.7129,
"step": 1540
},
{
"epoch": 0.75,
"grad_norm": 0.25061744451522827,
"learning_rate": 4.957849606741338e-05,
"loss": 0.6816,
"step": 1550
},
{
"epoch": 0.75,
"grad_norm": 0.1888163685798645,
"learning_rate": 4.956298918299903e-05,
"loss": 0.6905,
"step": 1560
},
{
"epoch": 0.76,
"grad_norm": 0.1991235315799713,
"learning_rate": 4.954720469699938e-05,
"loss": 0.7258,
"step": 1570
},
{
"epoch": 0.76,
"grad_norm": 0.18779252469539642,
"learning_rate": 4.953114278780471e-05,
"loss": 0.6918,
"step": 1580
},
{
"epoch": 0.77,
"grad_norm": 0.21268677711486816,
"learning_rate": 4.951480363694061e-05,
"loss": 0.6995,
"step": 1590
},
{
"epoch": 0.77,
"grad_norm": 0.20242677628993988,
"learning_rate": 4.949818742906597e-05,
"loss": 0.7132,
"step": 1600
},
{
"epoch": 0.78,
"grad_norm": 0.2314455509185791,
"learning_rate": 4.9481294351970874e-05,
"loss": 0.6866,
"step": 1610
},
{
"epoch": 0.78,
"grad_norm": 0.19638539850711823,
"learning_rate": 4.9464124596574465e-05,
"loss": 0.6993,
"step": 1620
},
{
"epoch": 0.79,
"grad_norm": 0.185603529214859,
"learning_rate": 4.944667835692281e-05,
"loss": 0.6971,
"step": 1630
},
{
"epoch": 0.79,
"grad_norm": 0.1594298928976059,
"learning_rate": 4.942895583018669e-05,
"loss": 0.7258,
"step": 1640
},
{
"epoch": 0.8,
"grad_norm": 0.21844029426574707,
"learning_rate": 4.9410957216659374e-05,
"loss": 0.7022,
"step": 1650
},
{
"epoch": 0.8,
"grad_norm": 0.19146642088890076,
"learning_rate": 4.9392682719754364e-05,
"loss": 0.6954,
"step": 1660
},
{
"epoch": 0.81,
"grad_norm": 0.20112347602844238,
"learning_rate": 4.9374132546003096e-05,
"loss": 0.7017,
"step": 1670
},
{
"epoch": 0.81,
"grad_norm": 0.22593769431114197,
"learning_rate": 4.935530690505259e-05,
"loss": 0.7397,
"step": 1680
},
{
"epoch": 0.82,
"grad_norm": 0.2611504793167114,
"learning_rate": 4.933620600966311e-05,
"loss": 0.7048,
"step": 1690
},
{
"epoch": 0.82,
"grad_norm": 0.23995471000671387,
"learning_rate": 4.931683007570571e-05,
"loss": 0.709,
"step": 1700
},
{
"epoch": 0.83,
"grad_norm": 0.22278735041618347,
"learning_rate": 4.9297179322159856e-05,
"loss": 0.7116,
"step": 1710
},
{
"epoch": 0.83,
"grad_norm": 0.2096502035856247,
"learning_rate": 4.927725397111093e-05,
"loss": 0.7061,
"step": 1720
},
{
"epoch": 0.84,
"grad_norm": 0.22335517406463623,
"learning_rate": 4.9257054247747644e-05,
"loss": 0.6982,
"step": 1730
},
{
"epoch": 0.84,
"grad_norm": 0.18597687780857086,
"learning_rate": 4.923658038035965e-05,
"loss": 0.6927,
"step": 1740
},
{
"epoch": 0.85,
"grad_norm": 0.1971946507692337,
"learning_rate": 4.9215832600334816e-05,
"loss": 0.6768,
"step": 1750
},
{
"epoch": 0.85,
"grad_norm": 0.2101408839225769,
"learning_rate": 4.9194811142156674e-05,
"loss": 0.6933,
"step": 1760
},
{
"epoch": 0.86,
"grad_norm": 0.17352430522441864,
"learning_rate": 4.917351624340178e-05,
"loss": 0.6918,
"step": 1770
},
{
"epoch": 0.86,
"grad_norm": 0.18568973243236542,
"learning_rate": 4.915194814473699e-05,
"loss": 0.6711,
"step": 1780
},
{
"epoch": 0.87,
"grad_norm": 0.22991472482681274,
"learning_rate": 4.913010708991679e-05,
"loss": 0.6938,
"step": 1790
},
{
"epoch": 0.87,
"grad_norm": 0.201872780919075,
"learning_rate": 4.910799332578048e-05,
"loss": 0.7051,
"step": 1800
},
{
"epoch": 0.87,
"grad_norm": 0.2069864720106125,
"learning_rate": 4.908560710224943e-05,
"loss": 0.6983,
"step": 1810
},
{
"epoch": 0.88,
"grad_norm": 0.2075955718755722,
"learning_rate": 4.906294867232426e-05,
"loss": 0.6996,
"step": 1820
},
{
"epoch": 0.88,
"grad_norm": 0.19524317979812622,
"learning_rate": 4.904001829208194e-05,
"loss": 0.691,
"step": 1830
},
{
"epoch": 0.89,
"grad_norm": 0.1967206597328186,
"learning_rate": 4.901681622067293e-05,
"loss": 0.6964,
"step": 1840
},
{
"epoch": 0.89,
"grad_norm": 0.2195650041103363,
"learning_rate": 4.899334272031825e-05,
"loss": 0.6982,
"step": 1850
},
{
"epoch": 0.9,
"grad_norm": 0.23072399199008942,
"learning_rate": 4.8969598056306475e-05,
"loss": 0.6838,
"step": 1860
},
{
"epoch": 0.9,
"grad_norm": 0.1950792670249939,
"learning_rate": 4.89455824969908e-05,
"loss": 0.6984,
"step": 1870
},
{
"epoch": 0.91,
"grad_norm": 0.2289438396692276,
"learning_rate": 4.892129631378597e-05,
"loss": 0.7012,
"step": 1880
},
{
"epoch": 0.91,
"grad_norm": 0.22316494584083557,
"learning_rate": 4.8896739781165215e-05,
"loss": 0.6904,
"step": 1890
},
{
"epoch": 0.92,
"grad_norm": 0.24066634476184845,
"learning_rate": 4.8871913176657135e-05,
"loss": 0.6974,
"step": 1900
},
{
"epoch": 0.92,
"grad_norm": 0.18922199308872223,
"learning_rate": 4.8846816780842606e-05,
"loss": 0.686,
"step": 1910
},
{
"epoch": 0.93,
"grad_norm": 0.2425553798675537,
"learning_rate": 4.882145087735158e-05,
"loss": 0.715,
"step": 1920
},
{
"epoch": 0.93,
"grad_norm": 0.2267230898141861,
"learning_rate": 4.879581575285988e-05,
"loss": 0.698,
"step": 1930
},
{
"epoch": 0.94,
"grad_norm": 0.22119645774364471,
"learning_rate": 4.876991169708592e-05,
"loss": 0.6942,
"step": 1940
},
{
"epoch": 0.94,
"grad_norm": 0.22854161262512207,
"learning_rate": 4.874373900278756e-05,
"loss": 0.6912,
"step": 1950
},
{
"epoch": 0.95,
"grad_norm": 0.1927136778831482,
"learning_rate": 4.871729796575863e-05,
"loss": 0.6829,
"step": 1960
},
{
"epoch": 0.95,
"grad_norm": 0.2238394170999527,
"learning_rate": 4.869058888482571e-05,
"loss": 0.6831,
"step": 1970
},
{
"epoch": 0.96,
"grad_norm": 0.17643775045871735,
"learning_rate": 4.86636120618447e-05,
"loss": 0.6875,
"step": 1980
},
{
"epoch": 0.96,
"grad_norm": 0.2161882072687149,
"learning_rate": 4.863636780169742e-05,
"loss": 0.6893,
"step": 1990
},
{
"epoch": 0.97,
"grad_norm": 0.20689886808395386,
"learning_rate": 4.860885641228815e-05,
"loss": 0.6721,
"step": 2000
},
{
"epoch": 0.97,
"eval_loss": 0.701475203037262,
"eval_runtime": 43.016,
"eval_samples_per_second": 46.494,
"eval_steps_per_second": 0.372,
"step": 2000
},
{
"epoch": 0.97,
"grad_norm": 0.18866655230522156,
"learning_rate": 4.8581078204540196e-05,
"loss": 0.6941,
"step": 2010
},
{
"epoch": 0.98,
"grad_norm": 0.20633092522621155,
"learning_rate": 4.855303349239231e-05,
"loss": 0.6943,
"step": 2020
},
{
"epoch": 0.98,
"grad_norm": 0.2335944026708603,
"learning_rate": 4.85247225927952e-05,
"loss": 0.6986,
"step": 2030
},
{
"epoch": 0.99,
"grad_norm": 0.18950659036636353,
"learning_rate": 4.8496145825707905e-05,
"loss": 0.6894,
"step": 2040
},
{
"epoch": 0.99,
"grad_norm": 0.22117875516414642,
"learning_rate": 4.8467303514094204e-05,
"loss": 0.6956,
"step": 2050
},
{
"epoch": 1.0,
"grad_norm": 0.23478396236896515,
"learning_rate": 4.8438195983918967e-05,
"loss": 0.6919,
"step": 2060
},
{
"epoch": 1.0,
"grad_norm": 0.24754443764686584,
"learning_rate": 4.8408823564144454e-05,
"loss": 0.6904,
"step": 2070
},
{
"epoch": 1.01,
"grad_norm": 0.19141903519630432,
"learning_rate": 4.837918658672661e-05,
"loss": 0.6914,
"step": 2080
},
{
"epoch": 1.01,
"grad_norm": 0.21387813985347748,
"learning_rate": 4.834928538661131e-05,
"loss": 0.6965,
"step": 2090
},
{
"epoch": 1.01,
"grad_norm": 0.20495550334453583,
"learning_rate": 4.831912030173058e-05,
"loss": 0.6949,
"step": 2100
},
{
"epoch": 1.02,
"grad_norm": 0.35174649953842163,
"learning_rate": 4.8288691672998765e-05,
"loss": 0.6813,
"step": 2110
},
{
"epoch": 1.02,
"grad_norm": 0.2461618334054947,
"learning_rate": 4.825799984430867e-05,
"loss": 0.6687,
"step": 2120
},
{
"epoch": 1.03,
"grad_norm": 0.20039792358875275,
"learning_rate": 4.8227045162527714e-05,
"loss": 0.6737,
"step": 2130
},
{
"epoch": 1.03,
"grad_norm": 0.19774436950683594,
"learning_rate": 4.8195827977493955e-05,
"loss": 0.6815,
"step": 2140
},
{
"epoch": 1.04,
"grad_norm": 0.21472251415252686,
"learning_rate": 4.8164348642012194e-05,
"loss": 0.6957,
"step": 2150
},
{
"epoch": 1.04,
"grad_norm": 0.1990274041891098,
"learning_rate": 4.813260751184992e-05,
"loss": 0.6916,
"step": 2160
},
{
"epoch": 1.05,
"grad_norm": 0.20387563109397888,
"learning_rate": 4.810060494573335e-05,
"loss": 0.6814,
"step": 2170
},
{
"epoch": 1.05,
"grad_norm": 0.1792985051870346,
"learning_rate": 4.806834130534336e-05,
"loss": 0.698,
"step": 2180
},
{
"epoch": 1.06,
"grad_norm": 0.22255714237689972,
"learning_rate": 4.803581695531134e-05,
"loss": 0.6765,
"step": 2190
},
{
"epoch": 1.06,
"grad_norm": 0.19549421966075897,
"learning_rate": 4.8003032263215185e-05,
"loss": 0.6867,
"step": 2200
},
{
"epoch": 1.07,
"grad_norm": 0.1929677277803421,
"learning_rate": 4.796998759957504e-05,
"loss": 0.6935,
"step": 2210
},
{
"epoch": 1.07,
"grad_norm": 0.22705228626728058,
"learning_rate": 4.793668333784915e-05,
"loss": 0.6897,
"step": 2220
},
{
"epoch": 1.08,
"grad_norm": 0.21798522770404816,
"learning_rate": 4.790311985442966e-05,
"loss": 0.6694,
"step": 2230
},
{
"epoch": 1.08,
"grad_norm": 0.19095788896083832,
"learning_rate": 4.7869297528638315e-05,
"loss": 0.6699,
"step": 2240
},
{
"epoch": 1.09,
"grad_norm": 0.23002374172210693,
"learning_rate": 4.7835216742722225e-05,
"loss": 0.6807,
"step": 2250
},
{
"epoch": 1.09,
"grad_norm": 0.2107260823249817,
"learning_rate": 4.780087788184947e-05,
"loss": 0.6855,
"step": 2260
},
{
"epoch": 1.1,
"grad_norm": 0.23853424191474915,
"learning_rate": 4.776628133410487e-05,
"loss": 0.6946,
"step": 2270
},
{
"epoch": 1.1,
"grad_norm": 0.25372380018234253,
"learning_rate": 4.7731427490485455e-05,
"loss": 0.6724,
"step": 2280
},
{
"epoch": 1.11,
"grad_norm": 0.182255819439888,
"learning_rate": 4.7696316744896146e-05,
"loss": 0.6976,
"step": 2290
},
{
"epoch": 1.11,
"grad_norm": 0.20137862861156464,
"learning_rate": 4.7660949494145276e-05,
"loss": 0.6853,
"step": 2300
},
{
"epoch": 1.12,
"grad_norm": 0.19899927079677582,
"learning_rate": 4.7625326137940106e-05,
"loss": 0.6696,
"step": 2310
},
{
"epoch": 1.12,
"grad_norm": 0.24147948622703552,
"learning_rate": 4.758944707888228e-05,
"loss": 0.6615,
"step": 2320
},
{
"epoch": 1.13,
"grad_norm": 0.19519510865211487,
"learning_rate": 4.7553312722463305e-05,
"loss": 0.685,
"step": 2330
},
{
"epoch": 1.13,
"grad_norm": 0.20128916203975677,
"learning_rate": 4.751692347705998e-05,
"loss": 0.6829,
"step": 2340
},
{
"epoch": 1.14,
"grad_norm": 0.16112156212329865,
"learning_rate": 4.748027975392976e-05,
"loss": 0.6714,
"step": 2350
},
{
"epoch": 1.14,
"grad_norm": 0.2659178674221039,
"learning_rate": 4.744338196720608e-05,
"loss": 0.6834,
"step": 2360
},
{
"epoch": 1.15,
"grad_norm": 0.20477250218391418,
"learning_rate": 4.740623053389374e-05,
"loss": 0.6664,
"step": 2370
},
{
"epoch": 1.15,
"grad_norm": 0.18069259822368622,
"learning_rate": 4.7368825873864154e-05,
"loss": 0.669,
"step": 2380
},
{
"epoch": 1.16,
"grad_norm": 0.2146090418100357,
"learning_rate": 4.733116840985058e-05,
"loss": 0.6836,
"step": 2390
},
{
"epoch": 1.16,
"grad_norm": 0.20656266808509827,
"learning_rate": 4.729325856744341e-05,
"loss": 0.6818,
"step": 2400
},
{
"epoch": 1.16,
"grad_norm": 0.1764717549085617,
"learning_rate": 4.725509677508528e-05,
"loss": 0.7012,
"step": 2410
},
{
"epoch": 1.17,
"grad_norm": 0.19059453904628754,
"learning_rate": 4.721668346406631e-05,
"loss": 0.6885,
"step": 2420
},
{
"epoch": 1.17,
"grad_norm": 0.1867195963859558,
"learning_rate": 4.7178019068519165e-05,
"loss": 0.6704,
"step": 2430
},
{
"epoch": 1.18,
"grad_norm": 0.19844096899032593,
"learning_rate": 4.713910402541416e-05,
"loss": 0.6746,
"step": 2440
},
{
"epoch": 1.18,
"grad_norm": 0.21426193416118622,
"learning_rate": 4.709993877455436e-05,
"loss": 0.6627,
"step": 2450
},
{
"epoch": 1.19,
"grad_norm": 0.19326262176036835,
"learning_rate": 4.706052375857058e-05,
"loss": 0.6652,
"step": 2460
},
{
"epoch": 1.19,
"grad_norm": 0.229305237531662,
"learning_rate": 4.7020859422916365e-05,
"loss": 0.6746,
"step": 2470
},
{
"epoch": 1.2,
"grad_norm": 0.20060321688652039,
"learning_rate": 4.698094621586299e-05,
"loss": 0.6658,
"step": 2480
},
{
"epoch": 1.2,
"grad_norm": 0.21455462276935577,
"learning_rate": 4.694078458849438e-05,
"loss": 0.6966,
"step": 2490
},
{
"epoch": 1.21,
"grad_norm": 0.18379837274551392,
"learning_rate": 4.690037499470202e-05,
"loss": 0.6753,
"step": 2500
},
{
"epoch": 1.21,
"eval_loss": 0.6940956711769104,
"eval_runtime": 42.7889,
"eval_samples_per_second": 46.741,
"eval_steps_per_second": 0.374,
"step": 2500
}
],
"logging_steps": 10,
"max_steps": 10345,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 6.328494039171596e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}