End of training

last-checkpoint/adapter_config.json

@@ -31,10 +31,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "o_proj",
+    "q_proj",
     "v_proj",
-    "q_proj"
+    "o_proj",
+    "k_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f22e14a21fbc01535d8b474a37f7897261e0e9d44ac6c70f5ae6599b1e461d7
+oid sha256:52217ccaabbb2c9a520ff6c890c15d98aa47a4c28fad51432c095d328a2a29be
 size 14764744

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2b171d33aa6f98342456fab464bee21a9699e54d6bfd82c63c9af6ea6adbb40
+oid sha256:a6af5e95ef3e872afad1bdf1fdf3ea5f5155675c6586239ad660ccd39c37347e
 size 29614667

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38930553ed957b1463a36a308942c5610692c94a54bf3656a9d7406bbfa9c1e1
+oid sha256:a82aeb7d859423fae1de432598feb0f08060628848416302157961cd81d904dc
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:497da62cc96f57ba96cd0fc26ae3953441e4ad7c8b6f524685215da682c18652
+oid sha256:51933555f7d7492da069489fcbf900fea3e97204e288093912746efcc0733e6e
 size 14917

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0861073ade655d14fb72097d362c2e813ed5d5faa50345ce5bc8f77fc935cfd8
+oid sha256:7d223d319f4b8402bb6d092355111dac11a4bb47d6e23f14a8f4c5bc3c5b9c8f
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa80ce769d9e6a0fc0b7fe734e59120bbbef8e392f0c0a33bd722e77bb1ee770
+oid sha256:dd465d73b04f6c59912627375bc7905b039595a268513d7eb113da343a1d81e2
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
+  "epoch": 0.8267665676269216,
   "eval_steps": 100,
-  "global_step": 5806,
+  "global_step": 4800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5336,1127 +5336,6 @@
       "eval_samples_per_second": 5.161,
       "eval_steps_per_second": 1.291,
       "step": 4800
-    },
-    {
-      "entropy": 0.857286661863327,
-      "epoch": 0.8284889979761444,
-      "grad_norm": 0.4545460641384125,
-      "learning_rate": 3.434378229417844e-06,
-      "loss": 0.8539424896240234,
-      "mean_token_accuracy": 0.8060515314340592,
-      "num_tokens": 136995.0,
-      "step": 4810
-    },
-    {
-      "entropy": 0.8996703952550889,
-      "epoch": 0.8302114283253671,
-      "grad_norm": 0.43648988008499146,
-      "learning_rate": 3.39993110575267e-06,
-      "loss": 0.8872912406921387,
-      "mean_token_accuracy": 0.8003117755055428,
-      "num_tokens": 269371.0,
-      "step": 4820
-    },
-    {
-      "entropy": 0.897844985127449,
-      "epoch": 0.8319338586745898,
-      "grad_norm": 0.405977338552475,
-      "learning_rate": 3.365483982087496e-06,
-      "loss": 0.8877438545227051,
-      "mean_token_accuracy": 0.8004682302474976,
-      "num_tokens": 407343.0,
-      "step": 4830
-    },
-    {
-      "entropy": 0.8919987097382546,
-      "epoch": 0.8336562890238126,
-      "grad_norm": 0.43606001138687134,
-      "learning_rate": 3.3310368584223218e-06,
-      "loss": 0.8915199279785156,
-      "mean_token_accuracy": 0.8018717259168625,
-      "num_tokens": 543192.0,
-      "step": 4840
-    },
-    {
-      "entropy": 0.8664562180638313,
-      "epoch": 0.8353787193730353,
-      "grad_norm": 0.4047588109970093,
-      "learning_rate": 3.296589734757148e-06,
-      "loss": 0.8543147087097168,
-      "mean_token_accuracy": 0.8059070542454719,
-      "num_tokens": 678711.0,
-      "step": 4850
-    },
-    {
-      "entropy": 0.9019272044301033,
-      "epoch": 0.8371011497222581,
-      "grad_norm": 0.40798190236091614,
-      "learning_rate": 3.262142611091974e-06,
-      "loss": 0.901366138458252,
-      "mean_token_accuracy": 0.7967334732413291,
-      "num_tokens": 815193.0,
-      "step": 4860
-    },
-    {
-      "entropy": 0.8930811554193496,
-      "epoch": 0.8388235800714808,
-      "grad_norm": 0.44522613286972046,
-      "learning_rate": 3.2276954874268003e-06,
-      "loss": 0.8780466079711914,
-      "mean_token_accuracy": 0.7989543616771698,
-      "num_tokens": 951512.0,
-      "step": 4870
-    },
-    {
-      "entropy": 0.897969713807106,
-      "epoch": 0.8405460104207036,
-      "grad_norm": 0.4132952094078064,
-      "learning_rate": 3.193248363761626e-06,
-      "loss": 0.89752197265625,
-      "mean_token_accuracy": 0.7986764028668404,
-      "num_tokens": 1091226.0,
-      "step": 4880
-    },
-    {
-      "entropy": 0.8763776004314423,
-      "epoch": 0.8422684407699264,
-      "grad_norm": 0.464271605014801,
-      "learning_rate": 3.1588012400964525e-06,
-      "loss": 0.866519546508789,
-      "mean_token_accuracy": 0.8051836535334587,
-      "num_tokens": 1230839.0,
-      "step": 4890
-    },
-    {
-      "entropy": 0.8817414686083793,
-      "epoch": 0.8439908711191492,
-      "grad_norm": 0.40158164501190186,
-      "learning_rate": 3.1243541164312784e-06,
-      "loss": 0.8720556259155273,
-      "mean_token_accuracy": 0.8009096115827561,
-      "num_tokens": 1371321.0,
-      "step": 4900
-    },
-    {
-      "epoch": 0.8439908711191492,
-      "eval_entropy": 0.8905009640159186,
-      "eval_loss": 0.886862576007843,
-      "eval_mean_token_accuracy": 0.8004815642430033,
-      "eval_num_tokens": 1371321.0,
-      "eval_runtime": 2214.5743,
-      "eval_samples_per_second": 4.66,
-      "eval_steps_per_second": 1.165,
-      "step": 4900
-    },
-    {
-      "entropy": 0.8792809963226318,
-      "epoch": 0.8457133014683719,
-      "grad_norm": 0.45575663447380066,
-      "learning_rate": 3.0899069927661042e-06,
-      "loss": 0.8713504791259765,
-      "mean_token_accuracy": 0.8019704148173332,
-      "num_tokens": 1508027.0,
-      "step": 4910
-    },
-    {
-      "entropy": 0.852638853341341,
-      "epoch": 0.8474357318175946,
-      "grad_norm": 0.5255561470985413,
-      "learning_rate": 3.05545986910093e-06,
-      "loss": 0.8388608932495117,
-      "mean_token_accuracy": 0.808800981938839,
-      "num_tokens": 1644684.0,
-      "step": 4920
-    },
-    {
-      "entropy": 0.852977229654789,
-      "epoch": 0.8491581621668174,
-      "grad_norm": 0.43369248509407043,
-      "learning_rate": 3.0210127454357564e-06,
-      "loss": 0.8463091850280762,
-      "mean_token_accuracy": 0.8071771740913392,
-      "num_tokens": 1782176.0,
-      "step": 4930
-    },
-    {
-      "entropy": 0.8610281966626644,
-      "epoch": 0.8508805925160401,
-      "grad_norm": 0.44929254055023193,
-      "learning_rate": 2.9865656217705823e-06,
-      "loss": 0.8570366859436035,
-      "mean_token_accuracy": 0.8059208214282989,
-      "num_tokens": 1920144.0,
-      "step": 4940
-    },
-    {
-      "entropy": 0.8712201908230781,
-      "epoch": 0.8526030228652629,
-      "grad_norm": 0.4670601487159729,
-      "learning_rate": 2.9521184981054086e-06,
-      "loss": 0.8666061401367188,
-      "mean_token_accuracy": 0.8054374232888222,
-      "num_tokens": 2059336.0,
-      "step": 4950
-    },
-    {
-      "entropy": 0.8756440788507461,
-      "epoch": 0.8543254532144856,
-      "grad_norm": 0.41233476996421814,
-      "learning_rate": 2.9176713744402345e-06,
-      "loss": 0.862638282775879,
-      "mean_token_accuracy": 0.8034965068101882,
-      "num_tokens": 2201332.0,
-      "step": 4960
-    },
-    {
-      "entropy": 0.9280423223972321,
-      "epoch": 0.8560478835637084,
-      "grad_norm": 0.37263938784599304,
-      "learning_rate": 2.883224250775061e-06,
-      "loss": 0.923769760131836,
-      "mean_token_accuracy": 0.7912818253040313,
-      "num_tokens": 2339368.0,
-      "step": 4970
-    },
-    {
-      "entropy": 0.8748306304216384,
-      "epoch": 0.8577703139129311,
-      "grad_norm": 0.47100237011909485,
-      "learning_rate": 2.8487771271098863e-06,
-      "loss": 0.8737285614013672,
-      "mean_token_accuracy": 0.8047029227018356,
-      "num_tokens": 2476367.0,
-      "step": 4980
-    },
-    {
-      "entropy": 0.9129098683595658,
-      "epoch": 0.8594927442621539,
-      "grad_norm": 0.528985321521759,
-      "learning_rate": 2.8143300034447126e-06,
-      "loss": 0.9065395355224609,
-      "mean_token_accuracy": 0.7945742771029473,
-      "num_tokens": 2613684.0,
-      "step": 4990
-    },
-    {
-      "entropy": 0.8374424375593662,
-      "epoch": 0.8612151746113766,
-      "grad_norm": 0.5343524217605591,
-      "learning_rate": 2.7798828797795385e-06,
-      "loss": 0.8323655128479004,
-      "mean_token_accuracy": 0.8117164567112922,
-      "num_tokens": 2748511.0,
-      "step": 5000
-    },
-    {
-      "epoch": 0.8612151746113766,
-      "eval_entropy": 0.8905347423907262,
-      "eval_loss": 0.886751651763916,
-      "eval_mean_token_accuracy": 0.8005248703313678,
-      "eval_num_tokens": 2748511.0,
-      "eval_runtime": 2213.3653,
-      "eval_samples_per_second": 4.663,
-      "eval_steps_per_second": 1.166,
-      "step": 5000
-    },
-    {
-      "entropy": 0.9117981031537056,
-      "epoch": 0.8629376049605995,
-      "grad_norm": 0.4343831539154053,
-      "learning_rate": 2.745435756114365e-06,
-      "loss": 0.9022266387939453,
-      "mean_token_accuracy": 0.7972292229533195,
-      "num_tokens": 2880947.0,
-      "step": 5010
-    },
-    {
-      "entropy": 0.9056061550974845,
-      "epoch": 0.8646600353098222,
-      "grad_norm": 0.578609824180603,
-      "learning_rate": 2.7109886324491907e-06,
-      "loss": 0.8953428268432617,
-      "mean_token_accuracy": 0.7994224324822425,
-      "num_tokens": 3018280.0,
-      "step": 5020
-    },
-    {
-      "entropy": 0.9164850175380707,
-      "epoch": 0.8663824656590449,
-      "grad_norm": 0.4623558223247528,
-      "learning_rate": 2.676541508784017e-06,
-      "loss": 0.9010392189025879,
-      "mean_token_accuracy": 0.7940018206834794,
-      "num_tokens": 3151878.0,
-      "step": 5030
-    },
-    {
-      "entropy": 0.9213888213038445,
-      "epoch": 0.8681048960082677,
-      "grad_norm": 0.4534718990325928,
-      "learning_rate": 2.642094385118843e-06,
-      "loss": 0.9162769317626953,
-      "mean_token_accuracy": 0.7945156171917915,
-      "num_tokens": 3289251.0,
-      "step": 5040
-    },
-    {
-      "entropy": 0.9475837752223015,
-      "epoch": 0.8698273263574904,
-      "grad_norm": 0.41711097955703735,
-      "learning_rate": 2.607647261453669e-06,
-      "loss": 0.9355989456176758,
-      "mean_token_accuracy": 0.7877664595842362,
-      "num_tokens": 3422221.0,
-      "step": 5050
-    },
-    {
-      "entropy": 0.859785021841526,
-      "epoch": 0.8715497567067132,
-      "grad_norm": 0.41227588057518005,
-      "learning_rate": 2.5732001377884947e-06,
-      "loss": 0.8524208068847656,
-      "mean_token_accuracy": 0.806971238553524,
-      "num_tokens": 3562863.0,
-      "step": 5060
-    },
-    {
-      "entropy": 0.9029248341917991,
-      "epoch": 0.8732721870559359,
-      "grad_norm": 0.597358226776123,
-      "learning_rate": 2.5387530141233206e-06,
-      "loss": 0.8965885162353515,
-      "mean_token_accuracy": 0.7993085518479347,
-      "num_tokens": 3700006.0,
-      "step": 5070
-    },
-    {
-      "entropy": 0.8976858526468277,
-      "epoch": 0.8749946174051587,
-      "grad_norm": 0.49744129180908203,
-      "learning_rate": 2.504305890458147e-06,
-      "loss": 0.8925239562988281,
-      "mean_token_accuracy": 0.7995666921138763,
-      "num_tokens": 3839690.0,
-      "step": 5080
-    },
-    {
-      "entropy": 0.9567562997341156,
-      "epoch": 0.8767170477543814,
-      "grad_norm": 0.4290732443332672,
-      "learning_rate": 2.469858766792973e-06,
-      "loss": 0.9521247863769531,
-      "mean_token_accuracy": 0.7859670892357826,
-      "num_tokens": 3972939.0,
-      "step": 5090
-    },
-    {
-      "entropy": 0.9057339891791344,
-      "epoch": 0.8784394781036042,
-      "grad_norm": 0.4539242684841156,
-      "learning_rate": 2.435411643127799e-06,
-      "loss": 0.89413480758667,
-      "mean_token_accuracy": 0.7966278985142707,
-      "num_tokens": 4106327.0,
-      "step": 5100
-    },
-    {
-      "epoch": 0.8784394781036042,
-      "eval_entropy": 0.889962231752624,
-      "eval_loss": 0.8865664005279541,
-      "eval_mean_token_accuracy": 0.8005476904811438,
-      "eval_num_tokens": 4106327.0,
-      "eval_runtime": 2216.2675,
-      "eval_samples_per_second": 4.657,
-      "eval_steps_per_second": 1.165,
-      "step": 5100
-    },
-    {
-      "entropy": 0.8694351434707641,
-      "epoch": 0.8801619084528269,
-      "grad_norm": 0.4723507761955261,
-      "learning_rate": 2.400964519462625e-06,
-      "loss": 0.8546782493591308,
-      "mean_token_accuracy": 0.8053493618965148,
-      "num_tokens": 4247505.0,
-      "step": 5110
-    },
-    {
-      "entropy": 0.8673742339015007,
-      "epoch": 0.8818843388020496,
-      "grad_norm": 0.4222690463066101,
-      "learning_rate": 2.3665173957974513e-06,
-      "loss": 0.8601169586181641,
-      "mean_token_accuracy": 0.8059160619974136,
-      "num_tokens": 4381996.0,
-      "step": 5120
-    },
-    {
-      "entropy": 0.8701221957802773,
-      "epoch": 0.8836067691512725,
-      "grad_norm": 0.365893691778183,
-      "learning_rate": 2.332070272132277e-06,
-      "loss": 0.8605199813842773,
-      "mean_token_accuracy": 0.8050397902727127,
-      "num_tokens": 4519936.0,
-      "step": 5130
-    },
-    {
-      "entropy": 0.9002679541707039,
-      "epoch": 0.8853291995004952,
-      "grad_norm": 0.41854503750801086,
-      "learning_rate": 2.2976231484671035e-06,
-      "loss": 0.8952171325683593,
-      "mean_token_accuracy": 0.7985835552215577,
-      "num_tokens": 4657396.0,
-      "step": 5140
-    },
-    {
-      "entropy": 0.8730725973844529,
-      "epoch": 0.887051629849718,
-      "grad_norm": 0.4184947907924652,
-      "learning_rate": 2.263176024801929e-06,
-      "loss": 0.874971866607666,
-      "mean_token_accuracy": 0.8019029393792152,
-      "num_tokens": 4793630.0,
-      "step": 5150
-    },
-    {
-      "entropy": 0.8665931671857834,
-      "epoch": 0.8887740601989407,
-      "grad_norm": 0.415822297334671,
-      "learning_rate": 2.2287289011367552e-06,
-      "loss": 0.8619186401367187,
-      "mean_token_accuracy": 0.8056416228413582,
-      "num_tokens": 4932856.0,
-      "step": 5160
-    },
-    {
-      "entropy": 0.8457090623676777,
-      "epoch": 0.8904964905481635,
-      "grad_norm": 0.5788813829421997,
-      "learning_rate": 2.1942817774715815e-06,
-      "loss": 0.8397070884704589,
-      "mean_token_accuracy": 0.8103364869952202,
-      "num_tokens": 5069138.0,
-      "step": 5170
-    },
-    {
-      "entropy": 0.8679139018058777,
-      "epoch": 0.8922189208973862,
-      "grad_norm": 0.4427867531776428,
-      "learning_rate": 2.1598346538064074e-06,
-      "loss": 0.8484273910522461,
-      "mean_token_accuracy": 0.8065137073397637,
-      "num_tokens": 5209109.0,
-      "step": 5180
-    },
-    {
-      "entropy": 0.8567999623715877,
-      "epoch": 0.893941351246609,
-      "grad_norm": 0.4930930435657501,
-      "learning_rate": 2.1253875301412333e-06,
-      "loss": 0.8500759124755859,
-      "mean_token_accuracy": 0.8062406674027442,
-      "num_tokens": 5347756.0,
-      "step": 5190
-    },
-    {
-      "entropy": 0.9153756201267242,
-      "epoch": 0.8956637815958317,
-      "grad_norm": 0.5475414395332336,
-      "learning_rate": 2.090940406476059e-06,
-      "loss": 0.913720703125,
-      "mean_token_accuracy": 0.798624849319458,
-      "num_tokens": 5484156.0,
-      "step": 5200
-    },
-    {
-      "epoch": 0.8956637815958317,
-      "eval_entropy": 0.8903410844865671,
-      "eval_loss": 0.8864830732345581,
-      "eval_mean_token_accuracy": 0.8005986783865671,
-      "eval_num_tokens": 5484156.0,
-      "eval_runtime": 2199.5544,
-      "eval_samples_per_second": 4.692,
-      "eval_steps_per_second": 1.173,
-      "step": 5200
-    },
-    {
-      "entropy": 0.8756707474589348,
-      "epoch": 0.8973862119450545,
-      "grad_norm": 0.4377118945121765,
-      "learning_rate": 2.0564932828108855e-06,
-      "loss": 0.8742427825927734,
-      "mean_token_accuracy": 0.8020416736602783,
-      "num_tokens": 5620110.0,
-      "step": 5210
-    },
-    {
-      "entropy": 0.9009493932127952,
-      "epoch": 0.8991086422942772,
-      "grad_norm": 0.385874480009079,
-      "learning_rate": 2.022046159145712e-06,
-      "loss": 0.9016963958740234,
-      "mean_token_accuracy": 0.7976742312312126,
-      "num_tokens": 5756362.0,
-      "step": 5220
-    },
-    {
-      "entropy": 0.8948132812976837,
-      "epoch": 0.9008310726434999,
-      "grad_norm": 0.4303382337093353,
-      "learning_rate": 1.9875990354805373e-06,
-      "loss": 0.8826349258422852,
-      "mean_token_accuracy": 0.7993564903736115,
-      "num_tokens": 5893634.0,
-      "step": 5230
-    },
-    {
-      "entropy": 0.9013346582651138,
-      "epoch": 0.9025535029927227,
-      "grad_norm": 0.43458035588264465,
-      "learning_rate": 1.9531519118153636e-06,
-      "loss": 0.8878791809082032,
-      "mean_token_accuracy": 0.7967622518539429,
-      "num_tokens": 6031714.0,
-      "step": 5240
-    },
-    {
-      "entropy": 0.900329303741455,
-      "epoch": 0.9042759333419454,
-      "grad_norm": 0.38891515135765076,
-      "learning_rate": 1.9187047881501895e-06,
-      "loss": 0.8943260192871094,
-      "mean_token_accuracy": 0.7962967336177826,
-      "num_tokens": 6167156.0,
-      "step": 5250
-    },
-    {
-      "entropy": 0.8946349762380124,
-      "epoch": 0.9059983636911683,
-      "grad_norm": 0.46829134225845337,
-      "learning_rate": 1.8842576644850158e-06,
-      "loss": 0.8836095809936524,
-      "mean_token_accuracy": 0.7979741424322129,
-      "num_tokens": 6304394.0,
-      "step": 5260
-    },
-    {
-      "entropy": 0.861349032074213,
-      "epoch": 0.907720794040391,
-      "grad_norm": 0.44049137830734253,
-      "learning_rate": 1.8498105408198417e-06,
-      "loss": 0.8538787841796875,
-      "mean_token_accuracy": 0.8079524412751198,
-      "num_tokens": 6445581.0,
-      "step": 5270
-    },
-    {
-      "entropy": 0.8150664329528808,
-      "epoch": 0.9094432243896138,
-      "grad_norm": 0.41160744428634644,
-      "learning_rate": 1.8153634171546678e-06,
-      "loss": 0.8064168930053711,
-      "mean_token_accuracy": 0.8146737694740296,
-      "num_tokens": 6586676.0,
-      "step": 5280
-    },
-    {
-      "entropy": 0.8552966974675655,
-      "epoch": 0.9111656547388365,
-      "grad_norm": 0.4296620190143585,
-      "learning_rate": 1.7809162934894939e-06,
-      "loss": 0.8459642410278321,
-      "mean_token_accuracy": 0.8074897557497025,
-      "num_tokens": 6725692.0,
-      "step": 5290
-    },
-    {
-      "entropy": 0.917242094874382,
-      "epoch": 0.9128880850880593,
-      "grad_norm": 0.4194020926952362,
-      "learning_rate": 1.7464691698243198e-06,
-      "loss": 0.9086408615112305,
-      "mean_token_accuracy": 0.7936558544635772,
-      "num_tokens": 6864077.0,
-      "step": 5300
-    },
-    {
-      "epoch": 0.9128880850880593,
-      "eval_entropy": 0.8895538462104376,
-      "eval_loss": 0.886420488357544,
-      "eval_mean_token_accuracy": 0.8006243253576344,
-      "eval_num_tokens": 6864077.0,
-      "eval_runtime": 2221.5521,
-      "eval_samples_per_second": 4.646,
-      "eval_steps_per_second": 1.162,
-      "step": 5300
-    },
-    {
-      "entropy": 0.8805374369025231,
-      "epoch": 0.914610515437282,
-      "grad_norm": 0.37022849917411804,
-      "learning_rate": 1.7120220461591459e-06,
-      "loss": 0.8718128204345703,
-      "mean_token_accuracy": 0.802679243683815,
-      "num_tokens": 6998242.0,
-      "step": 5310
-    },
-    {
-      "entropy": 0.9066009625792504,
-      "epoch": 0.9163329457865047,
-      "grad_norm": 0.6012776494026184,
-      "learning_rate": 1.677574922493972e-06,
-      "loss": 0.9060131072998047,
-      "mean_token_accuracy": 0.7947992786765099,
-      "num_tokens": 7136912.0,
-      "step": 5320
-    },
-    {
-      "entropy": 0.8422975018620491,
-      "epoch": 0.9180553761357275,
-      "grad_norm": 0.4443066716194153,
-      "learning_rate": 1.643127798828798e-06,
-      "loss": 0.8419797897338868,
-      "mean_token_accuracy": 0.8104378402233123,
-      "num_tokens": 7279298.0,
-      "step": 5330
-    },
-    {
-      "entropy": 0.8393626287579536,
-      "epoch": 0.9197778064849502,
-      "grad_norm": 0.42755192518234253,
-      "learning_rate": 1.608680675163624e-06,
-      "loss": 0.8286300659179687,
-      "mean_token_accuracy": 0.8120000720024109,
-      "num_tokens": 7416488.0,
-      "step": 5340
-    },
-    {
-      "entropy": 0.8612012237310409,
-      "epoch": 0.921500236834173,
-      "grad_norm": 0.42363837361335754,
-      "learning_rate": 1.57423355149845e-06,
-      "loss": 0.8535400390625,
-      "mean_token_accuracy": 0.8065411135554313,
-      "num_tokens": 7555188.0,
-      "step": 5350
-    },
-    {
-      "entropy": 0.8624228790402413,
-      "epoch": 0.9232226671833957,
-      "grad_norm": 0.4368661642074585,
-      "learning_rate": 1.5397864278332761e-06,
-      "loss": 0.8525060653686524,
-      "mean_token_accuracy": 0.8070851415395737,
-      "num_tokens": 7694113.0,
-      "step": 5360
-    },
-    {
-      "entropy": 0.89511788636446,
-      "epoch": 0.9249450975326186,
-      "grad_norm": 0.44434964656829834,
-      "learning_rate": 1.5053393041681022e-06,
-      "loss": 0.8826745033264161,
-      "mean_token_accuracy": 0.8017610549926758,
-      "num_tokens": 7831182.0,
-      "step": 5370
-    },
-    {
-      "entropy": 0.8900723055005073,
-      "epoch": 0.9266675278818413,
-      "grad_norm": 0.3896443545818329,
-      "learning_rate": 1.4708921805029281e-06,
-      "loss": 0.8831830978393554,
-      "mean_token_accuracy": 0.800842498242855,
-      "num_tokens": 7967772.0,
-      "step": 5380
-    },
-    {
-      "entropy": 0.8480780109763145,
-      "epoch": 0.9283899582310641,
-      "grad_norm": 0.47855135798454285,
-      "learning_rate": 1.4364450568377542e-06,
-      "loss": 0.8430942535400391,
-      "mean_token_accuracy": 0.8073634237051011,
-      "num_tokens": 8109699.0,
-      "step": 5390
-    },
-    {
-      "entropy": 0.8448813557624817,
-      "epoch": 0.9301123885802868,
-      "grad_norm": 0.40834853053092957,
-      "learning_rate": 1.4019979331725803e-06,
-      "loss": 0.8393930435180664,
-      "mean_token_accuracy": 0.8103328227996827,
-      "num_tokens": 8247058.0,
-      "step": 5400
-    },
-    {
-      "epoch": 0.9301123885802868,
-      "eval_entropy": 0.8897033281933194,
-      "eval_loss": 0.8863592147827148,
-      "eval_mean_token_accuracy": 0.8006240403592148,
-      "eval_num_tokens": 8247058.0,
-      "eval_runtime": 2208.4615,
-      "eval_samples_per_second": 4.673,
-      "eval_steps_per_second": 1.169,
-      "step": 5400
-    },
-    {
-      "entropy": 0.8857871651649475,
-      "epoch": 0.9318348189295096,
-      "grad_norm": 0.4397316575050354,
-      "learning_rate": 1.3675508095074064e-06,
-      "loss": 0.8728399276733398,
-      "mean_token_accuracy": 0.8018044486641884,
-      "num_tokens": 8384933.0,
-      "step": 5410
-    },
-    {
-      "entropy": 0.9105540379881859,
-      "epoch": 0.9335572492787323,
-      "grad_norm": 0.4545707702636719,
-      "learning_rate": 1.3331036858422323e-06,
-      "loss": 0.9085962295532226,
-      "mean_token_accuracy": 0.795919018983841,
-      "num_tokens": 8526364.0,
-      "step": 5420
-    },
-    {
-      "entropy": 0.8637060999870301,
-      "epoch": 0.935279679627955,
-      "grad_norm": 0.36590105295181274,
-      "learning_rate": 1.2986565621770584e-06,
-      "loss": 0.8607066154479981,
-      "mean_token_accuracy": 0.8060595214366912,
-      "num_tokens": 8664594.0,
-      "step": 5430
-    },
-    {
-      "entropy": 0.9462354958057404,
-      "epoch": 0.9370021099771778,
-      "grad_norm": 0.4243916869163513,
-      "learning_rate": 1.2642094385118845e-06,
-      "loss": 0.9325800895690918,
-      "mean_token_accuracy": 0.7894501313567162,
-      "num_tokens": 8798030.0,
-      "step": 5440
-    },
-    {
-      "entropy": 0.8978266701102257,
-      "epoch": 0.9387245403264005,
-      "grad_norm": 0.4084837734699249,
-      "learning_rate": 1.2297623148467104e-06,
-      "loss": 0.8851606369018554,
-      "mean_token_accuracy": 0.8009923383593559,
-      "num_tokens": 8937218.0,
-      "step": 5450
-    },
-    {
-      "entropy": 0.8797792851924896,
-      "epoch": 0.9404469706756233,
-      "grad_norm": 0.4102013409137726,
-      "learning_rate": 1.1953151911815365e-06,
-      "loss": 0.8767299652099609,
-      "mean_token_accuracy": 0.8016744539141655,
-      "num_tokens": 9075609.0,
-      "step": 5460
-    },
-    {
-      "entropy": 0.877577556669712,
-      "epoch": 0.942169401024846,
-      "grad_norm": 0.40450313687324524,
-      "learning_rate": 1.1608680675163626e-06,
-      "loss": 0.8800636291503906,
-      "mean_token_accuracy": 0.8021259069442749,
-      "num_tokens": 9215492.0,
-      "step": 5470
-    },
-    {
-      "entropy": 0.8801092639565468,
-      "epoch": 0.9438918313740688,
-      "grad_norm": 0.4468408226966858,
-      "learning_rate": 1.1264209438511885e-06,
-      "loss": 0.8776715278625489,
-      "mean_token_accuracy": 0.8040337398648262,
-      "num_tokens": 9355581.0,
-      "step": 5480
-    },
-    {
-      "entropy": 0.8833095215260982,
-      "epoch": 0.9456142617232915,
-      "grad_norm": 0.5369368195533752,
-      "learning_rate": 1.0919738201860146e-06,
-      "loss": 0.8716952323913574,
-      "mean_token_accuracy": 0.8020655795931816,
-      "num_tokens": 9492214.0,
-      "step": 5490
-    },
-    {
-      "entropy": 0.9180077627301216,
-      "epoch": 0.9473366920725144,
-      "grad_norm": 0.4173908829689026,
-      "learning_rate": 1.0575266965208405e-06,
-      "loss": 0.9027196884155273,
-      "mean_token_accuracy": 0.7945086941123009,
-      "num_tokens": 9630657.0,
-      "step": 5500
-    },
-    {
-      "epoch": 0.9473366920725144,
-      "eval_entropy": 0.89089541919515,
-      "eval_loss": 0.8862592577934265,
-      "eval_mean_token_accuracy": 0.8006462655479434,
-      "eval_num_tokens": 9630657.0,
-      "eval_runtime": 2217.3988,
-      "eval_samples_per_second": 4.655,
-      "eval_steps_per_second": 1.164,
-      "step": 5500
-    },
-    {
-      "entropy": 0.9045361965894699,
-      "epoch": 0.9490591224217371,
-      "grad_norm": 0.44494736194610596,
-      "learning_rate": 1.0230795728556668e-06,
-      "loss": 0.9016293525695801,
-      "mean_token_accuracy": 0.7991117894649505,
-      "num_tokens": 9765697.0,
-      "step": 5510
-    },
-    {
-      "entropy": 0.908899100124836,
-      "epoch": 0.9507815527709598,
-      "grad_norm": 0.4863620102405548,
-      "learning_rate": 9.886324491904926e-07,
-      "loss": 0.9006387710571289,
-      "mean_token_accuracy": 0.7961707264184952,
-      "num_tokens": 9902667.0,
-      "step": 5520
-    },
-    {
-      "entropy": 0.9102652281522751,
-      "epoch": 0.9525039831201826,
-      "grad_norm": 0.5516873598098755,
-      "learning_rate": 9.541853255253187e-07,
-      "loss": 0.8945722579956055,
-      "mean_token_accuracy": 0.7974153965711593,
-      "num_tokens": 10038391.0,
-      "step": 5530
-    },
-    {
-      "entropy": 0.8632800906896592,
-      "epoch": 0.9542264134694053,
-      "grad_norm": 0.420607328414917,
-      "learning_rate": 9.197382018601447e-07,
-      "loss": 0.8471515655517579,
-      "mean_token_accuracy": 0.8071994781494141,
-      "num_tokens": 10173686.0,
-      "step": 5540
-    },
-    {
-      "entropy": 0.885202006995678,
-      "epoch": 0.9559488438186281,
-      "grad_norm": 0.4669937789440155,
-      "learning_rate": 8.852910781949708e-07,
-      "loss": 0.886192512512207,
-      "mean_token_accuracy": 0.7994555085897446,
-      "num_tokens": 10311741.0,
-      "step": 5550
-    },
-    {
-      "entropy": 0.8559036374092102,
-      "epoch": 0.9576712741678508,
-      "grad_norm": 0.46770304441452026,
-      "learning_rate": 8.508439545297968e-07,
-      "loss": 0.8477782249450684,
-      "mean_token_accuracy": 0.8067097589373589,
-      "num_tokens": 10453099.0,
-      "step": 5560
-    },
-    {
-      "entropy": 0.9203044310212135,
-      "epoch": 0.9593937045170736,
-      "grad_norm": 0.4336691200733185,
-      "learning_rate": 8.163968308646229e-07,
-      "loss": 0.9105629920959473,
-      "mean_token_accuracy": 0.7944615721702576,
-      "num_tokens": 10591433.0,
-      "step": 5570
-    },
-    {
-      "entropy": 0.8992323987185955,
-      "epoch": 0.9611161348662963,
-      "grad_norm": 0.4164210557937622,
-      "learning_rate": 7.819497071994489e-07,
-      "loss": 0.8976916313171387,
-      "mean_token_accuracy": 0.7977223128080368,
-      "num_tokens": 10728991.0,
-      "step": 5580
-    },
-    {
-      "entropy": 0.8688705429434777,
-      "epoch": 0.9628385652155191,
-      "grad_norm": 0.4670332372188568,
-      "learning_rate": 7.47502583534275e-07,
-      "loss": 0.8597276687622071,
-      "mean_token_accuracy": 0.8061973169445992,
-      "num_tokens": 10867884.0,
-      "step": 5590
-    },
-    {
-      "entropy": 0.9053284600377083,
-      "epoch": 0.9645609955647418,
-      "grad_norm": 0.3938577175140381,
-      "learning_rate": 7.13055459869101e-07,
-      "loss": 0.8982341766357422,
-      "mean_token_accuracy": 0.7990050241351128,
-      "num_tokens": 11001110.0,
-      "step": 5600
-    },
-    {
-      "epoch": 0.9645609955647418,
-      "eval_entropy": 0.8901571277521785,
-      "eval_loss": 0.8862133026123047,
-      "eval_mean_token_accuracy": 0.8006129994590077,
-      "eval_num_tokens": 11001110.0,
-      "eval_runtime": 2221.0741,
-      "eval_samples_per_second": 4.647,
-      "eval_steps_per_second": 1.162,
-      "step": 5600
-    },
-    {
-      "entropy": 0.909729877114296,
-      "epoch": 0.9662834259139647,
-      "grad_norm": NaN,
-      "learning_rate": 6.78608336203927e-07,
-      "loss": 0.9095258712768555,
-      "mean_token_accuracy": 0.8012943014502525,
-      "num_tokens": 11139523.0,
-      "step": 5610
-    },
-    {
-      "entropy": 0.8407525554299354,
-      "epoch": 0.9680058562631874,
-      "grad_norm": 0.43725696206092834,
-      "learning_rate": 6.441612125387531e-07,
-      "loss": 0.8371444702148437,
-      "mean_token_accuracy": 0.810321681201458,
-      "num_tokens": 11276005.0,
-      "step": 5620
-    },
-    {
-      "entropy": 0.901355029642582,
-      "epoch": 0.9697282866124101,
-      "grad_norm": 0.5541648268699646,
-      "learning_rate": 6.097140888735791e-07,
-      "loss": 0.897189712524414,
-      "mean_token_accuracy": 0.7999485448002815,
-      "num_tokens": 11409561.0,
-      "step": 5630
-    },
-    {
-      "entropy": 0.9109830990433693,
-      "epoch": 0.9714507169616329,
-      "grad_norm": 0.5093613266944885,
-      "learning_rate": 5.752669652084052e-07,
-      "loss": 0.9014092445373535,
-      "mean_token_accuracy": 0.7963637366890908,
-      "num_tokens": 11545625.0,
-      "step": 5640
-    },
-    {
-      "entropy": 0.8920518733561039,
-      "epoch": 0.9731731473108556,
-      "grad_norm": 0.43631625175476074,
-      "learning_rate": 5.408198415432312e-07,
-      "loss": 0.8796001434326172,
-      "mean_token_accuracy": 0.7989365607500076,
-      "num_tokens": 11680063.0,
-      "step": 5650
-    },
-    {
-      "entropy": 0.9035694569349288,
-      "epoch": 0.9748955776600784,
-      "grad_norm": 0.40075695514678955,
-      "learning_rate": 5.063727178780573e-07,
-      "loss": 0.8951308250427246,
-      "mean_token_accuracy": 0.7964143276214599,
-      "num_tokens": 11819814.0,
-      "step": 5660
-    },
-    {
-      "entropy": 0.9101156964898109,
-      "epoch": 0.9766180080093011,
-      "grad_norm": 0.3921527564525604,
-      "learning_rate": 4.719255942128833e-07,
-      "loss": 0.9054186820983887,
-      "mean_token_accuracy": 0.7962990015745163,
-      "num_tokens": 11958312.0,
-      "step": 5670
-    },
-    {
-      "entropy": 0.8994546860456467,
-      "epoch": 0.9783404383585239,
-      "grad_norm": 0.4922078549861908,
-      "learning_rate": 4.3747847054770927e-07,
-      "loss": 0.8862943649291992,
-      "mean_token_accuracy": 0.7998211741447449,
-      "num_tokens": 12098750.0,
-      "step": 5680
-    },
-    {
-      "entropy": 0.8943607084453106,
-      "epoch": 0.9800628687077466,
-      "grad_norm": 0.35828521847724915,
-      "learning_rate": 4.030313468825353e-07,
-      "loss": 0.8841873168945312,
-      "mean_token_accuracy": 0.8006264850497246,
-      "num_tokens": 12238206.0,
-      "step": 5690
-    },
-    {
-      "entropy": 0.8610037744045258,
-      "epoch": 0.9817852990569694,
-      "grad_norm": 0.4048604965209961,
-      "learning_rate": 3.6858422321736136e-07,
-      "loss": 0.850681209564209,
-      "mean_token_accuracy": 0.8062293291091919,
-      "num_tokens": 12373855.0,
-      "step": 5700
-    },
-    {
-      "epoch": 0.9817852990569694,
-      "eval_entropy": 0.8902986454021464,
-      "eval_loss": 0.8861925601959229,
-      "eval_mean_token_accuracy": 0.8006364003210648,
-      "eval_num_tokens": 12373855.0,
-      "eval_runtime": 2198.4056,
-      "eval_samples_per_second": 4.695,
-      "eval_steps_per_second": 1.174,
-      "step": 5700
-    },
-    {
-      "entropy": 0.8190450638532638,
-      "epoch": 0.9835077294061921,
-      "grad_norm": 0.46052828431129456,
-      "learning_rate": 3.341370995521874e-07,
-      "loss": 0.8094842910766602,
-      "mean_token_accuracy": 0.8147952750325202,
-      "num_tokens": 12510961.0,
-      "step": 5710
-    },
-    {
-      "entropy": 0.9092415377497673,
-      "epoch": 0.9852301597554148,
-      "grad_norm": 0.46878325939178467,
-      "learning_rate": 2.9968997588701345e-07,
-      "loss": 0.8955598831176758,
-      "mean_token_accuracy": 0.7964127615094185,
-      "num_tokens": 12650748.0,
-      "step": 5720
-    },
-    {
-      "entropy": 0.8547648385167121,
-      "epoch": 0.9869525901046377,
-      "grad_norm": 0.47159335017204285,
-      "learning_rate": 2.652428522218395e-07,
-      "loss": 0.8555225372314453,
-      "mean_token_accuracy": 0.807997515797615,
-      "num_tokens": 12793599.0,
-      "step": 5730
-    },
-    {
-      "entropy": 0.8646257355809212,
-      "epoch": 0.9886750204538604,
-      "grad_norm": 0.48610877990722656,
-      "learning_rate": 2.3079572855666552e-07,
-      "loss": 0.8589855194091797,
-      "mean_token_accuracy": 0.806700712442398,
-      "num_tokens": 12932824.0,
-      "step": 5740
-    },
-    {
-      "entropy": 0.8452193960547447,
-      "epoch": 0.9903974508030832,
-      "grad_norm": 0.4377225637435913,
-      "learning_rate": 1.9634860489149156e-07,
-      "loss": 0.8359296798706055,
-      "mean_token_accuracy": 0.8110934183001518,
-      "num_tokens": 13071288.0,
-      "step": 5750
-    },
-    {
-      "entropy": 0.8705169402062893,
-      "epoch": 0.9921198811523059,
-      "grad_norm": 0.3535923659801483,
-      "learning_rate": 1.619014812263176e-07,
-      "loss": 0.8605812072753907,
-      "mean_token_accuracy": 0.804425074160099,
-      "num_tokens": 13209166.0,
-      "step": 5760
-    },
-    {
-      "entropy": 0.9036453366279602,
-      "epoch": 0.9938423115015287,
-      "grad_norm": 0.38716036081314087,
-      "learning_rate": 1.2745435756114365e-07,
-      "loss": 0.9031167030334473,
-      "mean_token_accuracy": 0.7980457425117493,
-      "num_tokens": 13344912.0,
-      "step": 5770
-    },
-    {
-      "entropy": 0.8937770411372185,
-      "epoch": 0.9955647418507514,
-      "grad_norm": 0.442220002412796,
-      "learning_rate": 9.300723389596969e-08,
-      "loss": 0.886353874206543,
-      "mean_token_accuracy": 0.7987162381410599,
-      "num_tokens": 13482642.0,
-      "step": 5780
-    },
-    {
-      "entropy": 0.8754376731812954,
-      "epoch": 0.9972871721999742,
-      "grad_norm": 0.3896678388118744,
-      "learning_rate": 5.856011023079573e-08,
-      "loss": 0.8634020805358886,
-      "mean_token_accuracy": 0.803331832587719,
-      "num_tokens": 13615820.0,
-      "step": 5790
-    },
-    {
-      "entropy": 0.891757769882679,
-      "epoch": 0.9990096025491969,
-      "grad_norm": 0.3823208808898926,
-      "learning_rate": 2.4112986565621773e-08,
-      "loss": 0.8841445922851563,
-      "mean_token_accuracy": 0.8023776292800904,
-      "num_tokens": 13750656.0,
-      "step": 5800
-    },
-    {
-      "epoch": 0.9990096025491969,
-      "eval_entropy": 0.8907162497113814,
-      "eval_loss": 0.8861898183822632,
-      "eval_mean_token_accuracy": 0.8006550548659512,
-      "eval_num_tokens": 13750656.0,
-      "eval_runtime": 2196.3123,
-      "eval_samples_per_second": 4.699,
-      "eval_steps_per_second": 1.175,
-      "step": 5800
-    },
-    {
-      "epoch": 1.0,
-      "eval_entropy": 0.8907080711504601,
-      "eval_loss": 0.8861844539642334,
-      "eval_mean_token_accuracy": 0.8006432617602077,
-      "eval_num_tokens": 13829843.0,
-      "eval_runtime": 2194.6346,
-      "eval_samples_per_second": 4.703,
-      "eval_steps_per_second": 1.176,
-      "step": 5806
     }
   ],
   "logging_steps": 10,
@@ -6471,12 +5350,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 9.53462095817474e+17,
+  "total_flos": 7.887012992552796e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null