first upload from kaggle

checkpoint-1182/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6cdac161c65a4b68dcf9ba007ba6f52dfacfd42d2554a1490e37b5030ba7dca9
+oid sha256:841f002f0348e373f18727bdd1a79c542747fbd6590d9f8f3330abfc78d84b62
 size 891644712

checkpoint-1182/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d191bf2d3cb3a44e8c60e2a9e3184e23cb1ebff95585e6e7013497a4b922c487
+oid sha256:061e75092c0432414c945a6b9b5a4e47e663d5dcf11823594c8e79a2ab4cbd2b
 size 1783444794

checkpoint-1182/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:835d84d51ea3b187a26482459bf2fecaf3fcb0bbe41bff40605191987b4fecf0
+oid sha256:6cfc0127b3222c5a66e96a5ce914651ccebce1c55569e992240485bf1c037abd
 size 14244

checkpoint-1182/trainer_state.json

@@ -11,876 +11,876 @@
   "log_history": [
     {
       "epoch": 0.050761421319796954,
-      "grad_norm": 2.660665988922119,
+      "grad_norm": 1.9596132040023804,
       "learning_rate": 4.961928934010153e-05,
-      "loss": 6.8339,
+      "loss": 3.0434,
       "step": 10
     },
     {
       "epoch": 0.10152284263959391,
-      "grad_norm": 1.533976674079895,
+      "grad_norm": 1.4552912712097168,
       "learning_rate": 4.919627749576988e-05,
-      "loss": 3.429,
+      "loss": 1.6984,
       "step": 20
     },
     {
       "epoch": 0.15228426395939088,
-      "grad_norm": 1.221410870552063,
+      "grad_norm": 1.4020640850067139,
       "learning_rate": 4.877326565143824e-05,
-      "loss": 2.5135,
+      "loss": 1.0141,
       "step": 30
     },
     {
       "epoch": 0.20304568527918782,
-      "grad_norm": 1.1820554733276367,
+      "grad_norm": 1.3040558099746704,
       "learning_rate": 4.83502538071066e-05,
-      "loss": 1.8569,
+      "loss": 0.6325,
       "step": 40
     },
     {
       "epoch": 0.25380710659898476,
-      "grad_norm": 1.0971927642822266,
+      "grad_norm": 0.9710696935653687,
       "learning_rate": 4.792724196277496e-05,
-      "loss": 1.4046,
+      "loss": 0.3562,
       "step": 50
     },
     {
       "epoch": 0.30456852791878175,
-      "grad_norm": 1.0558425188064575,
+      "grad_norm": 0.9529483914375305,
       "learning_rate": 4.750423011844332e-05,
-      "loss": 1.0788,
+      "loss": 0.2681,
       "step": 60
     },
     {
       "epoch": 0.3553299492385787,
-      "grad_norm": 0.9039331674575806,
+      "grad_norm": 0.9297605156898499,
       "learning_rate": 4.7081218274111674e-05,
-      "loss": 0.8565,
+      "loss": 0.1867,
       "step": 70
     },
     {
       "epoch": 0.40609137055837563,
-      "grad_norm": 0.9374755620956421,
+      "grad_norm": 0.6723515391349792,
       "learning_rate": 4.665820642978004e-05,
-      "loss": 0.6552,
+      "loss": 0.1557,
       "step": 80
     },
     {
       "epoch": 0.45685279187817257,
-      "grad_norm": 0.924466073513031,
+      "grad_norm": 0.5906422734260559,
       "learning_rate": 4.6235194585448395e-05,
-      "loss": 0.5584,
+      "loss": 0.1332,
       "step": 90
     },
     {
       "epoch": 0.5076142131979695,
-      "grad_norm": 0.7860057950019836,
+      "grad_norm": 0.562096357345581,
       "learning_rate": 4.5812182741116755e-05,
-      "loss": 0.4996,
+      "loss": 0.1113,
       "step": 100
     },
     {
       "epoch": 0.5583756345177665,
-      "grad_norm": 0.6625454425811768,
+      "grad_norm": 0.6856290102005005,
       "learning_rate": 4.538917089678511e-05,
-      "loss": 0.4086,
+      "loss": 0.0982,
       "step": 110
     },
     {
       "epoch": 0.6091370558375635,
-      "grad_norm": 0.6682830452919006,
+      "grad_norm": 0.3303697407245636,
       "learning_rate": 4.496615905245347e-05,
-      "loss": 0.3645,
+      "loss": 0.0794,
       "step": 120
     },
     {
       "epoch": 0.6598984771573604,
-      "grad_norm": 0.6779201626777649,
+      "grad_norm": 0.5941248536109924,
       "learning_rate": 4.454314720812183e-05,
-      "loss": 0.334,
+      "loss": 0.0799,
       "step": 130
     },
     {
       "epoch": 0.7106598984771574,
-      "grad_norm": 0.5805050730705261,
+      "grad_norm": 0.46145302057266235,
       "learning_rate": 4.412013536379019e-05,
-      "loss": 0.2835,
+      "loss": 0.0728,
       "step": 140
     },
     {
       "epoch": 0.7614213197969543,
-      "grad_norm": 0.6666281819343567,
+      "grad_norm": 0.5075628161430359,
       "learning_rate": 4.369712351945855e-05,
-      "loss": 0.2735,
+      "loss": 0.0705,
       "step": 150
     },
     {
       "epoch": 0.8121827411167513,
-      "grad_norm": 0.5769293308258057,
+      "grad_norm": 0.2965494394302368,
       "learning_rate": 4.32741116751269e-05,
-      "loss": 0.2341,
+      "loss": 0.0634,
       "step": 160
     },
     {
       "epoch": 0.8629441624365483,
-      "grad_norm": 0.5870254635810852,
+      "grad_norm": 0.3922906219959259,
       "learning_rate": 4.285109983079527e-05,
-      "loss": 0.2358,
+      "loss": 0.0599,
       "step": 170
     },
     {
       "epoch": 0.9137055837563451,
-      "grad_norm": 0.5556586384773254,
+      "grad_norm": 0.3413899540901184,
       "learning_rate": 4.242808798646362e-05,
-      "loss": 0.2271,
+      "loss": 0.0529,
       "step": 180
     },
     {
       "epoch": 0.9644670050761421,
-      "grad_norm": 0.5297289490699768,
+      "grad_norm": 0.37600159645080566,
       "learning_rate": 4.200507614213198e-05,
-      "loss": 0.2185,
+      "loss": 0.0548,
       "step": 190
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.11792419105768204,
-      "eval_runtime": 8.2824,
-      "eval_samples_per_second": 42.258,
-      "eval_steps_per_second": 2.656,
+      "eval_loss": 0.03613473102450371,
+      "eval_runtime": 6.8779,
+      "eval_samples_per_second": 50.887,
+      "eval_steps_per_second": 3.199,
       "step": 197
     },
     {
       "epoch": 1.015228426395939,
-      "grad_norm": 0.5531201958656311,
+      "grad_norm": 0.4854377508163452,
       "learning_rate": 4.1582064297800336e-05,
-      "loss": 0.1987,
+      "loss": 0.058,
       "step": 200
     },
     {
       "epoch": 1.0659898477157361,
-      "grad_norm": 0.5119745135307312,
+      "grad_norm": 0.39907264709472656,
       "learning_rate": 4.1159052453468696e-05,
-      "loss": 0.1823,
+      "loss": 0.0609,
       "step": 210
     },
     {
       "epoch": 1.116751269035533,
-      "grad_norm": 0.4964381158351898,
+      "grad_norm": 0.24890871345996857,
       "learning_rate": 4.073604060913706e-05,
-      "loss": 0.1783,
+      "loss": 0.0473,
       "step": 220
     },
     {
       "epoch": 1.16751269035533,
-      "grad_norm": 0.49353310465812683,
+      "grad_norm": 0.4353676736354828,
       "learning_rate": 4.0313028764805416e-05,
-      "loss": 0.1678,
+      "loss": 0.0513,
       "step": 230
     },
     {
       "epoch": 1.218274111675127,
-      "grad_norm": 0.44556906819343567,
+      "grad_norm": 0.38258448243141174,
       "learning_rate": 3.9890016920473777e-05,
-      "loss": 0.1704,
+      "loss": 0.0503,
       "step": 240
     },
     {
       "epoch": 1.2690355329949239,
-      "grad_norm": 0.4472508132457733,
+      "grad_norm": 0.3302125334739685,
       "learning_rate": 3.946700507614213e-05,
-      "loss": 0.1661,
+      "loss": 0.0478,
       "step": 250
     },
     {
       "epoch": 1.3197969543147208,
-      "grad_norm": 0.4651392996311188,
+      "grad_norm": 0.401644229888916,
       "learning_rate": 3.90439932318105e-05,
-      "loss": 0.1597,
+      "loss": 0.0457,
       "step": 260
     },
     {
       "epoch": 1.3705583756345177,
-      "grad_norm": 0.43394404649734497,
+      "grad_norm": 0.31225109100341797,
       "learning_rate": 3.862098138747885e-05,
-      "loss": 0.1557,
+      "loss": 0.0452,
       "step": 270
     },
     {
       "epoch": 1.4213197969543148,
-      "grad_norm": 0.40666043758392334,
+      "grad_norm": 0.30924656987190247,
       "learning_rate": 3.819796954314721e-05,
-      "loss": 0.1532,
+      "loss": 0.0428,
       "step": 280
     },
     {
       "epoch": 1.4720812182741116,
-      "grad_norm": 0.35698583722114563,
+      "grad_norm": 0.543154239654541,
       "learning_rate": 3.7774957698815564e-05,
-      "loss": 0.1423,
+      "loss": 0.048,
       "step": 290
     },
     {
       "epoch": 1.5228426395939088,
-      "grad_norm": 0.38902562856674194,
+      "grad_norm": 0.2982091009616852,
       "learning_rate": 3.735194585448393e-05,
-      "loss": 0.1446,
+      "loss": 0.0427,
       "step": 300
     },
     {
       "epoch": 1.5736040609137056,
-      "grad_norm": 0.43386396765708923,
+      "grad_norm": 0.3622360825538635,
       "learning_rate": 3.692893401015229e-05,
-      "loss": 0.1451,
+      "loss": 0.0431,
       "step": 310
     },
     {
       "epoch": 1.6243654822335025,
-      "grad_norm": 0.44021275639533997,
+      "grad_norm": 0.2379499226808548,
       "learning_rate": 3.6505922165820644e-05,
-      "loss": 0.1423,
+      "loss": 0.0408,
       "step": 320
     },
     {
       "epoch": 1.6751269035532994,
-      "grad_norm": 0.35672277212142944,
+      "grad_norm": 0.2724953889846802,
       "learning_rate": 3.6082910321489004e-05,
-      "loss": 0.1355,
+      "loss": 0.0419,
       "step": 330
     },
     {
       "epoch": 1.7258883248730963,
-      "grad_norm": 0.4129314124584198,
+      "grad_norm": 0.21542227268218994,
       "learning_rate": 3.565989847715736e-05,
-      "loss": 0.137,
+      "loss": 0.0439,
       "step": 340
     },
     {
       "epoch": 1.7766497461928934,
-      "grad_norm": 0.33145004510879517,
+      "grad_norm": 0.24891333281993866,
       "learning_rate": 3.5236886632825724e-05,
-      "loss": 0.1325,
+      "loss": 0.0393,
       "step": 350
     },
     {
       "epoch": 1.8274111675126905,
-      "grad_norm": 0.3832840323448181,
+      "grad_norm": 0.18472662568092346,
       "learning_rate": 3.481387478849408e-05,
-      "loss": 0.1299,
+      "loss": 0.0372,
       "step": 360
     },
     {
       "epoch": 1.8781725888324874,
-      "grad_norm": 0.4180988073348999,
+      "grad_norm": 0.1834375113248825,
       "learning_rate": 3.439086294416244e-05,
-      "loss": 0.1344,
+      "loss": 0.0383,
       "step": 370
     },
     {
       "epoch": 1.9289340101522843,
-      "grad_norm": 0.35554927587509155,
+      "grad_norm": 0.26916465163230896,
       "learning_rate": 3.396785109983079e-05,
-      "loss": 0.1207,
+      "loss": 0.0419,
       "step": 380
     },
     {
       "epoch": 1.9796954314720812,
-      "grad_norm": 0.3630838096141815,
+      "grad_norm": 0.2296602427959442,
       "learning_rate": 3.354483925549916e-05,
-      "loss": 0.1226,
+      "loss": 0.0391,
       "step": 390
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.08632376790046692,
-      "eval_runtime": 8.3339,
-      "eval_samples_per_second": 41.997,
-      "eval_steps_per_second": 2.64,
+      "eval_loss": 0.029596656560897827,
+      "eval_runtime": 6.9466,
+      "eval_samples_per_second": 50.384,
+      "eval_steps_per_second": 3.167,
       "step": 394
     },
     {
       "epoch": 2.030456852791878,
-      "grad_norm": 0.32654863595962524,
+      "grad_norm": 0.22394953668117523,
       "learning_rate": 3.312182741116752e-05,
-      "loss": 0.1246,
+      "loss": 0.0368,
       "step": 400
     },
     {
       "epoch": 2.081218274111675,
-      "grad_norm": 0.37500277161598206,
+      "grad_norm": 0.24742868542671204,
       "learning_rate": 3.269881556683587e-05,
-      "loss": 0.122,
+      "loss": 0.0417,
       "step": 410
     },
     {
       "epoch": 2.1319796954314723,
-      "grad_norm": 0.31452369689941406,
+      "grad_norm": 0.17821934819221497,
       "learning_rate": 3.227580372250423e-05,
-      "loss": 0.1195,
+      "loss": 0.039,
       "step": 420
     },
     {
       "epoch": 2.182741116751269,
-      "grad_norm": 0.29448235034942627,
+      "grad_norm": 0.17562909424304962,
       "learning_rate": 3.185279187817259e-05,
-      "loss": 0.1186,
+      "loss": 0.036,
       "step": 430
     },
     {
       "epoch": 2.233502538071066,
-      "grad_norm": 0.2858859598636627,
+      "grad_norm": 0.24495473504066467,
       "learning_rate": 3.142978003384095e-05,
-      "loss": 0.1169,
+      "loss": 0.038,
       "step": 440
     },
     {
       "epoch": 2.284263959390863,
-      "grad_norm": 0.32630932331085205,
+      "grad_norm": 0.21984700858592987,
       "learning_rate": 3.1006768189509306e-05,
-      "loss": 0.1161,
+      "loss": 0.0364,
       "step": 450
     },
     {
       "epoch": 2.33502538071066,
-      "grad_norm": 0.31745606660842896,
+      "grad_norm": 0.263046532869339,
       "learning_rate": 3.0583756345177666e-05,
-      "loss": 0.1175,
+      "loss": 0.0393,
       "step": 460
     },
     {
       "epoch": 2.3857868020304567,
-      "grad_norm": 0.359678715467453,
+      "grad_norm": 0.494204044342041,
       "learning_rate": 3.016074450084603e-05,
-      "loss": 0.112,
+      "loss": 0.0342,
       "step": 470
     },
     {
       "epoch": 2.436548223350254,
-      "grad_norm": 0.3830055594444275,
+      "grad_norm": 0.24457719922065735,
       "learning_rate": 2.9737732656514383e-05,
-      "loss": 0.1154,
+      "loss": 0.0371,
       "step": 480
     },
     {
       "epoch": 2.487309644670051,
-      "grad_norm": 0.35126161575317383,
+      "grad_norm": 0.2866905629634857,
       "learning_rate": 2.9314720812182743e-05,
-      "loss": 0.1159,
+      "loss": 0.0375,
       "step": 490
     },
     {
       "epoch": 2.5380710659898478,
-      "grad_norm": 0.32588812708854675,
+      "grad_norm": 0.1922035664319992,
       "learning_rate": 2.88917089678511e-05,
-      "loss": 0.112,
+      "loss": 0.0339,
       "step": 500
     },
     {
       "epoch": 2.5888324873096447,
-      "grad_norm": 0.349881649017334,
+      "grad_norm": 0.2251596301794052,
       "learning_rate": 2.846869712351946e-05,
-      "loss": 0.1088,
+      "loss": 0.0316,
       "step": 510
     },
     {
       "epoch": 2.6395939086294415,
-      "grad_norm": 0.287137508392334,
+      "grad_norm": 0.19956769049167633,
       "learning_rate": 2.8045685279187816e-05,
-      "loss": 0.1077,
+      "loss": 0.0367,
       "step": 520
     },
     {
       "epoch": 2.6903553299492384,
-      "grad_norm": 0.32511353492736816,
+      "grad_norm": 0.23161649703979492,
       "learning_rate": 2.7622673434856176e-05,
-      "loss": 0.1111,
+      "loss": 0.0335,
       "step": 530
     },
     {
       "epoch": 2.7411167512690353,
-      "grad_norm": 0.2147095799446106,
+      "grad_norm": 0.2735691964626312,
       "learning_rate": 2.7199661590524533e-05,
-      "loss": 0.1083,
+      "loss": 0.0367,
       "step": 540
     },
     {
       "epoch": 2.7918781725888326,
-      "grad_norm": 0.30077311396598816,
+      "grad_norm": 0.3856474757194519,
       "learning_rate": 2.6776649746192893e-05,
-      "loss": 0.1124,
+      "loss": 0.0362,
       "step": 550
     },
     {
       "epoch": 2.8426395939086295,
-      "grad_norm": 0.2887895703315735,
+      "grad_norm": 0.24519683420658112,
       "learning_rate": 2.6353637901861257e-05,
-      "loss": 0.1078,
+      "loss": 0.031,
       "step": 560
     },
     {
       "epoch": 2.8934010152284264,
-      "grad_norm": 0.28950536251068115,
+      "grad_norm": 0.12949654459953308,
       "learning_rate": 2.593062605752961e-05,
-      "loss": 0.106,
+      "loss": 0.032,
       "step": 570
     },
     {
       "epoch": 2.9441624365482233,
-      "grad_norm": 0.3079921007156372,
+      "grad_norm": 0.1476690173149109,
       "learning_rate": 2.5507614213197974e-05,
-      "loss": 0.1096,
+      "loss": 0.0351,
       "step": 580
     },
     {
       "epoch": 2.99492385786802,
-      "grad_norm": 0.23043136298656464,
+      "grad_norm": 0.24033169448375702,
       "learning_rate": 2.5084602368866327e-05,
-      "loss": 0.1034,
+      "loss": 0.0337,
       "step": 590
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.0816434845328331,
-      "eval_runtime": 8.6929,
-      "eval_samples_per_second": 40.263,
-      "eval_steps_per_second": 2.531,
+      "eval_loss": 0.02790662832558155,
+      "eval_runtime": 6.9258,
+      "eval_samples_per_second": 50.536,
+      "eval_steps_per_second": 3.177,
       "step": 591
     },
     {
       "epoch": 3.045685279187817,
-      "grad_norm": 0.2885898947715759,
+      "grad_norm": 0.25604188442230225,
       "learning_rate": 2.466159052453469e-05,
-      "loss": 0.1027,
+      "loss": 0.0339,
       "step": 600
     },
     {
       "epoch": 3.0964467005076144,
-      "grad_norm": 0.3026517927646637,
+      "grad_norm": 0.15198302268981934,
       "learning_rate": 2.4238578680203047e-05,
-      "loss": 0.1033,
+      "loss": 0.0316,
       "step": 610
     },
     {
       "epoch": 3.1472081218274113,
-      "grad_norm": 0.3037049472332001,
+      "grad_norm": 0.18943068385124207,
       "learning_rate": 2.3815566835871404e-05,
-      "loss": 0.1056,
+      "loss": 0.0302,
       "step": 620
     },
     {
       "epoch": 3.197969543147208,
-      "grad_norm": 0.2880629897117615,
+      "grad_norm": 0.23807291686534882,
       "learning_rate": 2.3392554991539764e-05,
-      "loss": 0.102,
+      "loss": 0.0338,
       "step": 630
     },
     {
       "epoch": 3.248730964467005,
-      "grad_norm": 0.23089496791362762,
+      "grad_norm": 0.2615777552127838,
       "learning_rate": 2.296954314720812e-05,
-      "loss": 0.1057,
+      "loss": 0.0291,
       "step": 640
     },
     {
       "epoch": 3.299492385786802,
-      "grad_norm": 0.30170634388923645,
+      "grad_norm": 0.20456817746162415,
       "learning_rate": 2.254653130287648e-05,
-      "loss": 0.1064,
+      "loss": 0.0331,
       "step": 650
     },
     {
       "epoch": 3.350253807106599,
-      "grad_norm": 0.21023182570934296,
+      "grad_norm": 0.29629555344581604,
       "learning_rate": 2.2123519458544838e-05,
-      "loss": 0.1054,
+      "loss": 0.0324,
       "step": 660
     },
     {
       "epoch": 3.401015228426396,
-      "grad_norm": 0.33246785402297974,
+      "grad_norm": 0.19070571660995483,
       "learning_rate": 2.17005076142132e-05,
-      "loss": 0.1053,
+      "loss": 0.0312,
       "step": 670
     },
     {
       "epoch": 3.451776649746193,
-      "grad_norm": 0.2597705125808716,
+      "grad_norm": 0.17927491664886475,
       "learning_rate": 2.1277495769881558e-05,
-      "loss": 0.1041,
+      "loss": 0.0331,
       "step": 680
     },
     {
       "epoch": 3.50253807106599,
-      "grad_norm": 0.28724995255470276,
+      "grad_norm": 0.16211186349391937,
       "learning_rate": 2.085448392554992e-05,
-      "loss": 0.1002,
+      "loss": 0.0324,
       "step": 690
     },
     {
       "epoch": 3.553299492385787,
-      "grad_norm": 0.23450523614883423,
+      "grad_norm": 0.13928809762001038,
       "learning_rate": 2.0431472081218275e-05,
-      "loss": 0.1011,
+      "loss": 0.0315,
       "step": 700
     },
     {
       "epoch": 3.6040609137055837,
-      "grad_norm": 0.3018731474876404,
+      "grad_norm": 0.2813867926597595,
       "learning_rate": 2.0008460236886635e-05,
-      "loss": 0.0987,
+      "loss": 0.03,
       "step": 710
     },
     {
       "epoch": 3.6548223350253806,
-      "grad_norm": 0.2904147207736969,
+      "grad_norm": 0.2689349353313446,
       "learning_rate": 1.9585448392554992e-05,
-      "loss": 0.1031,
+      "loss": 0.0333,
       "step": 720
     },
     {
       "epoch": 3.7055837563451774,
-      "grad_norm": 0.25796592235565186,
+      "grad_norm": 0.2879869043827057,
       "learning_rate": 1.916243654822335e-05,
-      "loss": 0.1032,
+      "loss": 0.035,
       "step": 730
     },
     {
       "epoch": 3.7563451776649748,
-      "grad_norm": 0.2625073194503784,
+      "grad_norm": 0.1638893336057663,
       "learning_rate": 1.873942470389171e-05,
-      "loss": 0.0975,
+      "loss": 0.0331,
       "step": 740
     },
     {
       "epoch": 3.8071065989847717,
-      "grad_norm": 0.2199327051639557,
+      "grad_norm": 0.08905433863401413,
       "learning_rate": 1.831641285956007e-05,
-      "loss": 0.1015,
+      "loss": 0.0291,
       "step": 750
     },
     {
       "epoch": 3.8578680203045685,
-      "grad_norm": 0.24463722109794617,
+      "grad_norm": 0.2221483290195465,
       "learning_rate": 1.789340101522843e-05,
-      "loss": 0.103,
+      "loss": 0.0334,
       "step": 760
     },
     {
       "epoch": 3.9086294416243654,
-      "grad_norm": 0.25082287192344666,
+      "grad_norm": 0.16910187900066376,
       "learning_rate": 1.7470389170896786e-05,
-      "loss": 0.0978,
+      "loss": 0.0331,
       "step": 770
     },
     {
       "epoch": 3.9593908629441623,
-      "grad_norm": 0.24737493693828583,
+      "grad_norm": 0.20653125643730164,
       "learning_rate": 1.7047377326565146e-05,
-      "loss": 0.1014,
+      "loss": 0.0328,
       "step": 780
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.0787789523601532,
-      "eval_runtime": 8.3689,
-      "eval_samples_per_second": 41.822,
-      "eval_steps_per_second": 2.629,
+      "eval_loss": 0.02677118219435215,
+      "eval_runtime": 6.9163,
+      "eval_samples_per_second": 50.605,
+      "eval_steps_per_second": 3.181,
       "step": 788
     },
     {
       "epoch": 4.01015228426396,
-      "grad_norm": 0.35116952657699585,
+      "grad_norm": 0.5460578203201294,
       "learning_rate": 1.6624365482233503e-05,
-      "loss": 0.1018,
+      "loss": 0.0317,
       "step": 790
     },
     {
       "epoch": 4.060913705583756,
-      "grad_norm": 0.258379191160202,
+      "grad_norm": 0.1273794323205948,
       "learning_rate": 1.6201353637901863e-05,
-      "loss": 0.1005,
+      "loss": 0.0324,
       "step": 800
     },
     {
       "epoch": 4.111675126903553,
-      "grad_norm": 0.25503265857696533,
+      "grad_norm": 0.2069994956254959,
       "learning_rate": 1.577834179357022e-05,
-      "loss": 0.0992,
+      "loss": 0.0328,
       "step": 810
     },
     {
       "epoch": 4.16243654822335,
-      "grad_norm": 0.23493419587612152,
+      "grad_norm": 0.13560791313648224,
       "learning_rate": 1.535532994923858e-05,
-      "loss": 0.1006,
+      "loss": 0.0289,
       "step": 820
     },
     {
       "epoch": 4.213197969543147,
-      "grad_norm": 0.23928825557231903,
+      "grad_norm": 0.13835355639457703,
       "learning_rate": 1.493231810490694e-05,
-      "loss": 0.1013,
+      "loss": 0.0285,
       "step": 830
     },
     {
       "epoch": 4.2639593908629445,
-      "grad_norm": 0.28521081805229187,
+      "grad_norm": 0.17146103084087372,
       "learning_rate": 1.4509306260575298e-05,
-      "loss": 0.1008,
+      "loss": 0.0328,
       "step": 840
     },
     {
       "epoch": 4.314720812182741,
-      "grad_norm": 0.2244846224784851,
+      "grad_norm": 0.25955504179000854,
       "learning_rate": 1.4086294416243657e-05,
-      "loss": 0.0978,
+      "loss": 0.0295,
       "step": 850
     },
     {
       "epoch": 4.365482233502538,
-      "grad_norm": 0.32869383692741394,
+      "grad_norm": 0.24718697369098663,
       "learning_rate": 1.3663282571912014e-05,
-      "loss": 0.0969,
+      "loss": 0.0307,
       "step": 860
     },
     {
       "epoch": 4.416243654822335,
-      "grad_norm": 0.23633809387683868,
+      "grad_norm": 0.12164635211229324,
       "learning_rate": 1.3240270727580372e-05,
-      "loss": 0.1013,
+      "loss": 0.0287,
       "step": 870
     },
     {
       "epoch": 4.467005076142132,
-      "grad_norm": 0.3110709488391876,
+      "grad_norm": 0.17382808029651642,
       "learning_rate": 1.281725888324873e-05,
-      "loss": 0.0937,
+      "loss": 0.0472,
       "step": 880
     },
     {
       "epoch": 4.517766497461929,
-      "grad_norm": 0.22026929259300232,
+      "grad_norm": 0.17402203381061554,
       "learning_rate": 1.239424703891709e-05,
-      "loss": 0.0982,
+      "loss": 0.0343,
       "step": 890
     },
     {
       "epoch": 4.568527918781726,
-      "grad_norm": 0.23756134510040283,
+      "grad_norm": 0.17245104908943176,
       "learning_rate": 1.1971235194585449e-05,
-      "loss": 0.0954,
+      "loss": 0.0318,
       "step": 900
     },
     {
       "epoch": 4.619289340101523,
-      "grad_norm": 0.24070966243743896,
+      "grad_norm": 0.1376132220029831,
       "learning_rate": 1.1548223350253808e-05,
-      "loss": 0.1013,
+      "loss": 0.0319,
       "step": 910
     },
     {
       "epoch": 4.67005076142132,
-      "grad_norm": 0.22478370368480682,
+      "grad_norm": 0.17528069019317627,
       "learning_rate": 1.1125211505922166e-05,
-      "loss": 0.0928,
+      "loss": 0.0302,
       "step": 920
     },
     {
       "epoch": 4.720812182741117,
-      "grad_norm": 0.19010470807552338,
+      "grad_norm": 0.2443544864654541,
       "learning_rate": 1.0702199661590526e-05,
-      "loss": 0.0986,
+      "loss": 0.0295,
       "step": 930
     },
     {
       "epoch": 4.771573604060913,
-      "grad_norm": 0.26357847452163696,
+      "grad_norm": 0.21152476966381073,
       "learning_rate": 1.0279187817258885e-05,
-      "loss": 0.0898,
+      "loss": 0.0331,
       "step": 940
     },
     {
       "epoch": 4.822335025380711,
-      "grad_norm": 0.2300788015127182,
+      "grad_norm": 0.13216163218021393,
       "learning_rate": 9.856175972927243e-06,
-      "loss": 0.0958,
+      "loss": 0.0283,
       "step": 950
     },
     {
       "epoch": 4.873096446700508,
-      "grad_norm": 0.25872883200645447,
+      "grad_norm": 0.1937057226896286,
       "learning_rate": 9.433164128595601e-06,
-      "loss": 0.0957,
+      "loss": 0.0285,
       "step": 960
     },
     {
       "epoch": 4.9238578680203045,
-      "grad_norm": 0.21223337948322296,
+      "grad_norm": 0.1196654811501503,
       "learning_rate": 9.01015228426396e-06,
-      "loss": 0.0992,
+      "loss": 0.0299,
       "step": 970
     },
     {
       "epoch": 4.974619289340102,
-      "grad_norm": 0.2555055320262909,
+      "grad_norm": 0.14108304679393768,
       "learning_rate": 8.587140439932318e-06,
-      "loss": 0.0953,
+      "loss": 0.0326,
       "step": 980
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.07795918732881546,
-      "eval_runtime": 8.314,
-      "eval_samples_per_second": 42.098,
-      "eval_steps_per_second": 2.646,
+      "eval_loss": 0.026293369010090828,
+      "eval_runtime": 7.2804,
+      "eval_samples_per_second": 48.074,
+      "eval_steps_per_second": 3.022,
       "step": 985
     },
     {
       "epoch": 5.025380710659898,
-      "grad_norm": 0.22976677119731903,
+      "grad_norm": 0.11325781047344208,
       "learning_rate": 8.164128595600677e-06,
-      "loss": 0.0938,
+      "loss": 0.0303,
       "step": 990
     },
     {
       "epoch": 5.0761421319796955,
-      "grad_norm": 0.21435701847076416,
+      "grad_norm": 0.1742030531167984,
       "learning_rate": 7.741116751269035e-06,
-      "loss": 0.0938,
+      "loss": 0.029,
       "step": 1000
     },
     {
       "epoch": 5.126903553299492,
-      "grad_norm": 0.22609326243400574,
+      "grad_norm": 0.19924026727676392,
       "learning_rate": 7.318104906937395e-06,
-      "loss": 0.0977,
+      "loss": 0.0271,
       "step": 1010
     },
     {
       "epoch": 5.177664974619289,
-      "grad_norm": 0.19491638243198395,
+      "grad_norm": 0.23700544238090515,
       "learning_rate": 6.895093062605754e-06,
-      "loss": 0.0958,
+      "loss": 0.0306,
       "step": 1020
     },
     {
       "epoch": 5.228426395939087,
-      "grad_norm": 0.23073458671569824,
+      "grad_norm": 0.12165335565805435,
       "learning_rate": 6.472081218274112e-06,
-      "loss": 0.0877,
+      "loss": 0.0318,
       "step": 1030
     },
     {
       "epoch": 5.279187817258883,
-      "grad_norm": 0.24976614117622375,
+      "grad_norm": 0.21364423632621765,
       "learning_rate": 6.049069373942471e-06,
-      "loss": 0.0939,
+      "loss": 0.03,
       "step": 1040
     },
     {
       "epoch": 5.32994923857868,
-      "grad_norm": 0.18394580483436584,
+      "grad_norm": 0.19045327603816986,
       "learning_rate": 5.626057529610829e-06,
-      "loss": 0.0907,
+      "loss": 0.0325,
       "step": 1050
     },
     {
       "epoch": 5.380710659898477,
-      "grad_norm": 0.24421803653240204,
+      "grad_norm": 0.10052906721830368,
       "learning_rate": 5.203045685279188e-06,
-      "loss": 0.0956,
+      "loss": 0.0278,
       "step": 1060
     },
     {
       "epoch": 5.431472081218274,
-      "grad_norm": 0.24214482307434082,
+      "grad_norm": 0.2044578194618225,
       "learning_rate": 4.780033840947547e-06,
-      "loss": 0.0989,
+      "loss": 0.0309,
       "step": 1070
     },
     {
       "epoch": 5.482233502538071,
-      "grad_norm": 0.24704593420028687,
+      "grad_norm": 0.19502834975719452,
       "learning_rate": 4.357021996615906e-06,
-      "loss": 0.0963,
+      "loss": 0.0318,
       "step": 1080
     },
     {
       "epoch": 5.532994923857868,
-      "grad_norm": 0.2930368483066559,
+      "grad_norm": 0.13834765553474426,
       "learning_rate": 3.934010152284264e-06,
-      "loss": 0.0919,
+      "loss": 0.0305,
       "step": 1090
     },
     {
       "epoch": 5.583756345177665,
-      "grad_norm": 0.17989520728588104,
+      "grad_norm": 0.19017720222473145,
       "learning_rate": 3.5109983079526226e-06,
-      "loss": 0.0965,
+      "loss": 0.0305,
       "step": 1100
     },
     {
       "epoch": 5.634517766497462,
-      "grad_norm": 0.19147488474845886,
+      "grad_norm": 0.14318296313285828,
       "learning_rate": 3.0879864636209815e-06,
-      "loss": 0.0923,
+      "loss": 0.0304,
       "step": 1110
     },
     {
       "epoch": 5.685279187817259,
-      "grad_norm": 0.2020110785961151,
+      "grad_norm": 0.13694196939468384,
       "learning_rate": 2.6649746192893404e-06,
-      "loss": 0.0968,
+      "loss": 0.0301,
       "step": 1120
     },
     {
       "epoch": 5.7360406091370555,
-      "grad_norm": 0.2407018393278122,
+      "grad_norm": 0.11877632886171341,
       "learning_rate": 2.241962774957699e-06,
-      "loss": 0.0956,
+      "loss": 0.0303,
       "step": 1130
     },
     {
       "epoch": 5.786802030456853,
-      "grad_norm": 0.3431318998336792,
+      "grad_norm": 0.1271430402994156,
       "learning_rate": 1.8189509306260577e-06,
-      "loss": 0.0911,
+      "loss": 0.0321,
       "step": 1140
     },
     {
       "epoch": 5.837563451776649,
-      "grad_norm": 0.26111626625061035,
+      "grad_norm": 0.1693529337644577,
       "learning_rate": 1.3959390862944163e-06,
-      "loss": 0.0944,
+      "loss": 0.0318,
       "step": 1150
     },
     {
       "epoch": 5.888324873096447,
-      "grad_norm": 0.2500187158584595,
+      "grad_norm": 0.1400621086359024,
       "learning_rate": 9.72927241962775e-07,
-      "loss": 0.0945,
+      "loss": 0.0273,
       "step": 1160
     },
     {
       "epoch": 5.939086294416244,
-      "grad_norm": 0.22575554251670837,
+      "grad_norm": 0.20201422274112701,
       "learning_rate": 5.499153976311337e-07,
-      "loss": 0.0983,
+      "loss": 0.0287,
       "step": 1170
     },
     {
       "epoch": 5.98984771573604,
-      "grad_norm": 0.19752255082130432,
+      "grad_norm": 0.12207765877246857,
       "learning_rate": 1.2690355329949238e-07,
-      "loss": 0.0967,
+      "loss": 0.0313,
       "step": 1180
     },
     {
       "epoch": 6.0,
-      "eval_loss": 0.07735521346330643,
-      "eval_runtime": 8.3449,
-      "eval_samples_per_second": 41.942,
-      "eval_steps_per_second": 2.636,
+      "eval_loss": 0.02633434534072876,
+      "eval_runtime": 6.8922,
+      "eval_samples_per_second": 50.782,
+      "eval_steps_per_second": 3.192,
       "step": 1182
     }
   ],
@@ -901,7 +901,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2580518529792000.0,
+  "total_flos": 2581201228677120.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6cdac161c65a4b68dcf9ba007ba6f52dfacfd42d2554a1490e37b5030ba7dca9
+oid sha256:841f002f0348e373f18727bdd1a79c542747fbd6590d9f8f3330abfc78d84b62
 size 891644712