Update fine-tuned model

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "up_proj",
-    "q_proj",
+    "v_proj",
     "gate_proj",
     "down_proj",
+    "o_proj",
     "k_proj",
-    "v_proj"
+    "up_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f6a49f69351456485e9eee6012dfa1780295775f8d7d1ac3267aded02fb47ea
+oid sha256:2d6a977183e505b764d0dfd8b152287fb444c9cf075dabe509af0cdc23a3a374
 size 159967880

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06728820a0d2910a7c148685fd6ca4a6f0ed059643904223033daa0932396a53
+oid sha256:dfa58f767b69164ca37825b684f8eff7d5bf21a6e04391029c45a229c5c64612
 size 852876198

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d02782d602efcdfe0b1c7a56f8520f6d57aed1961efb8e2e93c91f029291f316
+oid sha256:f8ad58ab3a10b5d635a4a58ef5dd2e25339fd79e7202a61585dca7b9458a71aa
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a60c7d771c1fd156acee762fba03c724cb41829a3f71df370ecd1d20b134982
+oid sha256:25c0243da09e051ffbb600413baac079966a139cd6939cbef7eb321283bd0767
 size 1064

trainer_state.json

@@ -1,718 +1,893 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.0,
+  "epoch": 5.0,
   "eval_steps": 500,
-  "global_step": 100,
+  "global_step": 125,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.04,
-      "grad_norm": 2.0647521018981934,
-      "learning_rate": 4e-05,
-      "loss": 1.1825,
+      "grad_norm": 1.5257668495178223,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 0.9817,
       "step": 1
     },
     {
       "epoch": 0.08,
-      "grad_norm": 1.9963816404342651,
-      "learning_rate": 8e-05,
-      "loss": 1.1448,
+      "grad_norm": 1.4205083847045898,
+      "learning_rate": 5.714285714285714e-05,
+      "loss": 0.9677,
       "step": 2
     },
     {
       "epoch": 0.12,
-      "grad_norm": 1.82085120677948,
-      "learning_rate": 0.00012,
-      "loss": 1.1246,
+      "grad_norm": 1.4500166177749634,
+      "learning_rate": 8.571428571428571e-05,
+      "loss": 0.9229,
       "step": 3
     },
     {
       "epoch": 0.16,
-      "grad_norm": 1.3797991275787354,
-      "learning_rate": 0.00016,
-      "loss": 0.9428,
+      "grad_norm": 1.1406153440475464,
+      "learning_rate": 0.00011428571428571428,
+      "loss": 0.8997,
       "step": 4
     },
     {
       "epoch": 0.2,
-      "grad_norm": 1.0062527656555176,
-      "learning_rate": 0.0002,
-      "loss": 0.8782,
+      "grad_norm": 0.9545271396636963,
+      "learning_rate": 0.00014285714285714287,
+      "loss": 0.8483,
       "step": 5
     },
     {
       "epoch": 0.24,
-      "grad_norm": 0.6854935884475708,
-      "learning_rate": 0.00019994532573409262,
-      "loss": 0.8182,
+      "grad_norm": 0.8234877586364746,
+      "learning_rate": 0.00017142857142857143,
+      "loss": 0.8351,
       "step": 6
     },
     {
       "epoch": 0.28,
-      "grad_norm": 0.6088404059410095,
-      "learning_rate": 0.00019978136272187747,
-      "loss": 0.7337,
+      "grad_norm": 0.6187587380409241,
+      "learning_rate": 0.0002,
+      "loss": 0.7575,
       "step": 7
     },
     {
       "epoch": 0.32,
-      "grad_norm": 0.5058557391166687,
-      "learning_rate": 0.00019950829025450114,
-      "loss": 0.7606,
+      "grad_norm": 0.4950176775455475,
+      "learning_rate": 0.00019996456111234527,
+      "loss": 0.7344,
       "step": 8
     },
     {
       "epoch": 0.36,
-      "grad_norm": 0.4383871853351593,
-      "learning_rate": 0.00019912640693269752,
-      "loss": 0.7934,
+      "grad_norm": 0.43441540002822876,
+      "learning_rate": 0.0001998582695676762,
+      "loss": 0.7035,
       "step": 9
     },
     {
       "epoch": 0.4,
-      "grad_norm": 0.3763003349304199,
-      "learning_rate": 0.00019863613034027224,
-      "loss": 0.7623,
+      "grad_norm": 0.4230976402759552,
+      "learning_rate": 0.000199681200703075,
+      "loss": 0.7492,
       "step": 10
     },
     {
       "epoch": 0.44,
-      "grad_norm": 0.3510531187057495,
-      "learning_rate": 0.00019803799658748094,
-      "loss": 0.7043,
+      "grad_norm": 0.3713449239730835,
+      "learning_rate": 0.00019943348002101371,
+      "loss": 0.6831,
       "step": 11
     },
     {
       "epoch": 0.48,
-      "grad_norm": 0.3534976541996002,
-      "learning_rate": 0.0001973326597248006,
-      "loss": 0.7702,
+      "grad_norm": 0.3475893437862396,
+      "learning_rate": 0.00019911528310040074,
+      "loss": 0.7153,
       "step": 12
     },
     {
       "epoch": 0.52,
-      "grad_norm": 0.33237403631210327,
-      "learning_rate": 0.00019652089102773488,
-      "loss": 0.7074,
+      "grad_norm": 0.3414737582206726,
+      "learning_rate": 0.00019872683547213446,
+      "loss": 0.6943,
       "step": 13
     },
     {
       "epoch": 0.56,
-      "grad_norm": 0.3145230710506439,
-      "learning_rate": 0.00019560357815343577,
-      "loss": 0.6778,
+      "grad_norm": 0.3290707767009735,
+      "learning_rate": 0.00019826841245925212,
+      "loss": 0.6698,
       "step": 14
     },
     {
       "epoch": 0.6,
-      "grad_norm": 0.32996928691864014,
-      "learning_rate": 0.00019458172417006347,
-      "loss": 0.7032,
+      "grad_norm": 0.3098256587982178,
+      "learning_rate": 0.00019774033898178667,
+      "loss": 0.6385,
       "step": 15
     },
     {
       "epoch": 0.64,
-      "grad_norm": 0.32944002747535706,
-      "learning_rate": 0.0001934564464599461,
-      "loss": 0.7088,
+      "grad_norm": 0.31097927689552307,
+      "learning_rate": 0.00019714298932647098,
+      "loss": 0.6694,
       "step": 16
     },
     {
       "epoch": 0.68,
-      "grad_norm": 0.3067278265953064,
-      "learning_rate": 0.00019222897549773848,
-      "loss": 0.6803,
+      "grad_norm": 0.32695478200912476,
+      "learning_rate": 0.0001964767868814516,
+      "loss": 0.6702,
       "step": 17
     },
     {
       "epoch": 0.72,
-      "grad_norm": 0.3252066671848297,
-      "learning_rate": 0.00019090065350491626,
-      "loss": 0.6947,
+      "grad_norm": 0.30651360750198364,
+      "learning_rate": 0.00019574220383620055,
+      "loss": 0.6172,
       "step": 18
     },
     {
       "epoch": 0.76,
-      "grad_norm": 0.31279104948043823,
-      "learning_rate": 0.00018947293298207635,
-      "loss": 0.6402,
+      "grad_norm": 0.31658628582954407,
+      "learning_rate": 0.00019493976084683813,
+      "loss": 0.6248,
       "step": 19
     },
     {
       "epoch": 0.8,
-      "grad_norm": 0.30892765522003174,
-      "learning_rate": 0.0001879473751206489,
-      "loss": 0.6519,
+      "grad_norm": 0.30691343545913696,
+      "learning_rate": 0.00019407002666710336,
+      "loss": 0.6281,
       "step": 20
     },
     {
       "epoch": 0.84,
-      "grad_norm": 0.29156580567359924,
-      "learning_rate": 0.00018632564809575742,
-      "loss": 0.6747,
+      "grad_norm": 0.29928600788116455,
+      "learning_rate": 0.00019313361774523385,
+      "loss": 0.6208,
       "step": 21
     },
     {
       "epoch": 0.88,
-      "grad_norm": 0.2854658365249634,
-      "learning_rate": 0.00018460952524209355,
-      "loss": 0.6556,
+      "grad_norm": 0.30110830068588257,
+      "learning_rate": 0.00019213119778704128,
+      "loss": 0.5995,
       "step": 22
     },
     {
       "epoch": 0.92,
-      "grad_norm": 0.2657647430896759,
-      "learning_rate": 0.00018280088311480201,
-      "loss": 0.6128,
+      "grad_norm": 0.2957627475261688,
+      "learning_rate": 0.00019106347728549135,
+      "loss": 0.6016,
       "step": 23
     },
     {
       "epoch": 0.96,
-      "grad_norm": 0.29050764441490173,
-      "learning_rate": 0.00018090169943749476,
-      "loss": 0.6674,
+      "grad_norm": 0.29468321800231934,
+      "learning_rate": 0.00018993121301712193,
+      "loss": 0.5997,
       "step": 24
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.29195520281791687,
-      "learning_rate": 0.00017891405093963938,
-      "loss": 0.63,
+      "grad_norm": 0.2975432276725769,
+      "learning_rate": 0.00018873520750565718,
+      "loss": 0.6106,
       "step": 25
     },
     {
       "epoch": 1.04,
-      "grad_norm": 0.30334123969078064,
-      "learning_rate": 0.00017684011108568592,
-      "loss": 0.5533,
+      "grad_norm": 0.31368955969810486,
+      "learning_rate": 0.00018747630845319612,
+      "loss": 0.4904,
       "step": 26
     },
     {
       "epoch": 1.08,
-      "grad_norm": 0.30424660444259644,
-      "learning_rate": 0.0001746821476984154,
-      "loss": 0.5549,
+      "grad_norm": 0.31520742177963257,
+      "learning_rate": 0.0001861554081393806,
+      "loss": 0.5047,
       "step": 27
     },
     {
       "epoch": 1.12,
-      "grad_norm": 0.28369948267936707,
-      "learning_rate": 0.00017244252047910892,
-      "loss": 0.5415,
+      "grad_norm": 0.3065387010574341,
+      "learning_rate": 0.0001847734427889671,
+      "loss": 0.4789,
       "step": 28
     },
     {
       "epoch": 1.16,
-      "grad_norm": 0.2778087258338928,
-      "learning_rate": 0.00017012367842724887,
-      "loss": 0.5382,
+      "grad_norm": 0.2928008437156677,
+      "learning_rate": 0.0001833313919082515,
+      "loss": 0.4898,
       "step": 29
     },
     {
       "epoch": 1.2,
-      "grad_norm": 0.2843495011329651,
-      "learning_rate": 0.00016772815716257412,
-      "loss": 0.5388,
+      "grad_norm": 0.2972809672355652,
+      "learning_rate": 0.0001818302775908169,
+      "loss": 0.4904,
       "step": 30
     },
     {
       "epoch": 1.24,
-      "grad_norm": 0.28178870677948,
-      "learning_rate": 0.00016525857615241687,
-      "loss": 0.5126,
+      "grad_norm": 0.2960495948791504,
+      "learning_rate": 0.00018027116379309638,
+      "loss": 0.4629,
       "step": 31
     },
     {
       "epoch": 1.28,
-      "grad_norm": 0.28643131256103516,
-      "learning_rate": 0.0001627176358473537,
-      "loss": 0.5167,
+      "grad_norm": 0.28314685821533203,
+      "learning_rate": 0.00017865515558026428,
+      "loss": 0.4483,
       "step": 32
     },
     {
       "epoch": 1.32,
-      "grad_norm": 0.28843289613723755,
-      "learning_rate": 0.00016010811472830252,
-      "loss": 0.5099,
+      "grad_norm": 0.2923860549926758,
+      "learning_rate": 0.00017698339834299061,
+      "loss": 0.4602,
       "step": 33
     },
     {
       "epoch": 1.3599999999999999,
-      "grad_norm": 0.28687819838523865,
-      "learning_rate": 0.00015743286626829437,
-      "loss": 0.5185,
+      "grad_norm": 0.2865493893623352,
+      "learning_rate": 0.00017525707698561385,
+      "loss": 0.4565,
       "step": 34
     },
     {
       "epoch": 1.4,
-      "grad_norm": 0.28858163952827454,
-      "learning_rate": 0.00015469481581224272,
-      "loss": 0.5279,
+      "grad_norm": 0.29644760489463806,
+      "learning_rate": 0.00017347741508630672,
+      "loss": 0.4491,
       "step": 35
     },
     {
       "epoch": 1.44,
-      "grad_norm": 0.2918362021446228,
-      "learning_rate": 0.00015189695737812152,
-      "loss": 0.4788,
+      "grad_norm": 0.2860000729560852,
+      "learning_rate": 0.00017164567402983152,
+      "loss": 0.444,
       "step": 36
     },
     {
       "epoch": 1.48,
-      "grad_norm": 0.291130930185318,
-      "learning_rate": 0.00014904235038305083,
-      "loss": 0.4856,
+      "grad_norm": 0.2956458330154419,
+      "learning_rate": 0.0001697631521134985,
+      "loss": 0.464,
       "step": 37
     },
     {
       "epoch": 1.52,
-      "grad_norm": 0.2943766415119171,
-      "learning_rate": 0.0001461341162978688,
-      "loss": 0.4615,
+      "grad_norm": 0.30596640706062317,
+      "learning_rate": 0.00016783118362696163,
+      "loss": 0.4596,
       "step": 38
     },
     {
       "epoch": 1.56,
-      "grad_norm": 0.297374427318573,
-      "learning_rate": 0.00014317543523384928,
-      "loss": 0.4583,
+      "grad_norm": 0.31658679246902466,
+      "learning_rate": 0.00016585113790650388,
+      "loss": 0.4184,
       "step": 39
     },
     {
       "epoch": 1.6,
-      "grad_norm": 0.30772677063941956,
-      "learning_rate": 0.00014016954246529696,
-      "loss": 0.4859,
+      "grad_norm": 0.3067062497138977,
+      "learning_rate": 0.00016382441836448202,
+      "loss": 0.3952,
       "step": 40
     },
     {
       "epoch": 1.6400000000000001,
-      "grad_norm": 0.29820892214775085,
-      "learning_rate": 0.00013711972489182208,
-      "loss": 0.4768,
+      "grad_norm": 0.30973076820373535,
+      "learning_rate": 0.0001617524614946192,
+      "loss": 0.4109,
       "step": 41
     },
     {
       "epoch": 1.6800000000000002,
-      "grad_norm": 0.29821571707725525,
-      "learning_rate": 0.00013402931744416433,
-      "loss": 0.4891,
+      "grad_norm": 0.30597272515296936,
+      "learning_rate": 0.00015963673585385016,
+      "loss": 0.3853,
       "step": 42
     },
     {
       "epoch": 1.72,
-      "grad_norm": 0.2966136634349823,
-      "learning_rate": 0.00013090169943749476,
-      "loss": 0.4794,
+      "grad_norm": 0.3238050639629364,
+      "learning_rate": 0.0001574787410214407,
+      "loss": 0.4074,
       "step": 43
     },
     {
       "epoch": 1.76,
-      "grad_norm": 0.3031865358352661,
-      "learning_rate": 0.00012774029087618446,
-      "loss": 0.4853,
+      "grad_norm": 0.32239633798599243,
+      "learning_rate": 0.00015528000653611935,
+      "loss": 0.421,
       "step": 44
     },
     {
       "epoch": 1.8,
-      "grad_norm": 0.2918758690357208,
-      "learning_rate": 0.00012454854871407994,
-      "loss": 0.4523,
+      "grad_norm": 0.322157084941864,
+      "learning_rate": 0.00015304209081197425,
+      "loss": 0.4156,
       "step": 45
     },
     {
       "epoch": 1.8399999999999999,
-      "grad_norm": 0.3020419180393219,
-      "learning_rate": 0.0001213299630743747,
-      "loss": 0.4668,
+      "grad_norm": 0.32195115089416504,
+      "learning_rate": 0.000150766580033884,
+      "loss": 0.3955,
       "step": 46
     },
     {
       "epoch": 1.88,
-      "grad_norm": 0.30048614740371704,
-      "learning_rate": 0.000118088053433211,
-      "loss": 0.4468,
+      "grad_norm": 0.312031626701355,
+      "learning_rate": 0.00014845508703326504,
+      "loss": 0.3814,
       "step": 47
     },
     {
       "epoch": 1.92,
-      "grad_norm": 0.3022134304046631,
-      "learning_rate": 0.0001148263647711842,
-      "loss": 0.4636,
+      "grad_norm": 0.31998252868652344,
+      "learning_rate": 0.0001461092501449326,
+      "loss": 0.4039,
       "step": 48
     },
     {
       "epoch": 1.96,
-      "grad_norm": 0.3102932572364807,
-      "learning_rate": 0.00011154846369695863,
-      "loss": 0.4762,
+      "grad_norm": 0.32456570863723755,
+      "learning_rate": 0.00014373073204588556,
+      "loss": 0.3963,
       "step": 49
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.29303210973739624,
-      "learning_rate": 0.00010825793454723325,
-      "loss": 0.4109,
+      "grad_norm": 0.32449814677238464,
+      "learning_rate": 0.00014132121857683783,
+      "loss": 0.3845,
       "step": 50
     },
     {
       "epoch": 2.04,
-      "grad_norm": 0.29739847779273987,
-      "learning_rate": 0.00010495837546732224,
-      "loss": 0.3787,
+      "grad_norm": 0.31362128257751465,
+      "learning_rate": 0.00013888241754733208,
+      "loss": 0.3138,
       "step": 51
     },
     {
       "epoch": 2.08,
-      "grad_norm": 0.29793456196784973,
-      "learning_rate": 0.00010165339447663587,
-      "loss": 0.3708,
+      "grad_norm": 0.30638328194618225,
+      "learning_rate": 0.00013641605752528224,
+      "loss": 0.317,
       "step": 52
     },
     {
       "epoch": 2.12,
-      "grad_norm": 0.31656354665756226,
-      "learning_rate": 9.834660552336415e-05,
-      "loss": 0.3855,
+      "grad_norm": 0.32126715779304504,
+      "learning_rate": 0.00013392388661180303,
+      "loss": 0.3004,
       "step": 53
     },
     {
       "epoch": 2.16,
-      "grad_norm": 0.30371928215026855,
-      "learning_rate": 9.504162453267777e-05,
-      "loss": 0.3873,
+      "grad_norm": 0.28879204392433167,
+      "learning_rate": 0.0001314076712021949,
+      "loss": 0.2997,
       "step": 54
     },
     {
       "epoch": 2.2,
-      "grad_norm": 0.31064578890800476,
-      "learning_rate": 9.174206545276677e-05,
-      "loss": 0.3595,
+      "grad_norm": 0.29814431071281433,
+      "learning_rate": 0.0001288691947339621,
+      "loss": 0.2934,
       "step": 55
     },
     {
       "epoch": 2.24,
-      "grad_norm": 0.29833751916885376,
-      "learning_rate": 8.845153630304139e-05,
-      "loss": 0.366,
+      "grad_norm": 0.28737854957580566,
+      "learning_rate": 0.00012631025642275212,
+      "loss": 0.3185,
       "step": 56
     },
     {
       "epoch": 2.2800000000000002,
-      "grad_norm": 0.3072323501110077,
-      "learning_rate": 8.517363522881579e-05,
-      "loss": 0.3621,
+      "grad_norm": 0.30604204535484314,
+      "learning_rate": 0.0001237326699871115,
+      "loss": 0.3059,
       "step": 57
     },
     {
       "epoch": 2.32,
-      "grad_norm": 0.29597151279449463,
-      "learning_rate": 8.191194656678904e-05,
-      "loss": 0.3519,
+      "grad_norm": 0.2846826910972595,
+      "learning_rate": 0.00012113826236296244,
+      "loss": 0.2779,
       "step": 58
     },
     {
       "epoch": 2.36,
-      "grad_norm": 0.3137385845184326,
-      "learning_rate": 7.867003692562534e-05,
-      "loss": 0.3519,
+      "grad_norm": 0.3072626292705536,
+      "learning_rate": 0.00011852887240871145,
+      "loss": 0.2958,
       "step": 59
     },
     {
       "epoch": 2.4,
-      "grad_norm": 0.3161297142505646,
-      "learning_rate": 7.54514512859201e-05,
-      "loss": 0.3502,
+      "grad_norm": 0.30152225494384766,
+      "learning_rate": 0.00011590634960190721,
+      "loss": 0.276,
       "step": 60
     },
     {
       "epoch": 2.44,
-      "grad_norm": 0.3286161720752716,
-      "learning_rate": 7.225970912381556e-05,
-      "loss": 0.3518,
+      "grad_norm": 0.31316888332366943,
+      "learning_rate": 0.00011327255272837221,
+      "loss": 0.2863,
       "step": 61
     },
     {
       "epoch": 2.48,
-      "grad_norm": 0.3324306905269623,
-      "learning_rate": 6.909830056250527e-05,
-      "loss": 0.357,
+      "grad_norm": 0.33458268642425537,
+      "learning_rate": 0.00011062934856473655,
+      "loss": 0.2999,
       "step": 62
     },
     {
       "epoch": 2.52,
-      "grad_norm": 0.3510464131832123,
-      "learning_rate": 6.59706825558357e-05,
-      "loss": 0.379,
+      "grad_norm": 0.3459411561489105,
+      "learning_rate": 0.00010797861055530831,
+      "loss": 0.2861,
       "step": 63
     },
     {
       "epoch": 2.56,
-      "grad_norm": 0.3358787000179291,
-      "learning_rate": 6.28802751081779e-05,
-      "loss": 0.3399,
+      "grad_norm": 0.34091514348983765,
+      "learning_rate": 0.00010532221748421787,
+      "loss": 0.2697,
       "step": 64
     },
     {
       "epoch": 2.6,
-      "grad_norm": 0.330174058675766,
-      "learning_rate": 5.983045753470308e-05,
-      "loss": 0.3465,
+      "grad_norm": 0.3289962708950043,
+      "learning_rate": 0.00010266205214377748,
+      "loss": 0.2722,
       "step": 65
     },
     {
       "epoch": 2.64,
-      "grad_norm": 0.31485724449157715,
-      "learning_rate": 5.6824564766150726e-05,
-      "loss": 0.3249,
+      "grad_norm": 0.324381023645401,
+      "learning_rate": 0.0001,
+      "loss": 0.2864,
       "step": 66
     },
     {
       "epoch": 2.68,
-      "grad_norm": 0.3314495384693146,
-      "learning_rate": 5.386588370213124e-05,
-      "loss": 0.3565,
+      "grad_norm": 0.34481576085090637,
+      "learning_rate": 9.733794785622253e-05,
+      "loss": 0.3002,
       "step": 67
     },
     {
       "epoch": 2.7199999999999998,
-      "grad_norm": 0.334344744682312,
-      "learning_rate": 5.095764961694922e-05,
-      "loss": 0.344,
+      "grad_norm": 0.33097216486930847,
+      "learning_rate": 9.467778251578217e-05,
+      "loss": 0.2915,
       "step": 68
     },
     {
       "epoch": 2.76,
-      "grad_norm": 0.330770343542099,
-      "learning_rate": 4.810304262187852e-05,
-      "loss": 0.3424,
+      "grad_norm": 0.3260320723056793,
+      "learning_rate": 9.202138944469168e-05,
+      "loss": 0.2962,
       "step": 69
     },
     {
       "epoch": 2.8,
-      "grad_norm": 0.32071828842163086,
-      "learning_rate": 4.530518418775733e-05,
-      "loss": 0.3396,
+      "grad_norm": 0.3088409900665283,
+      "learning_rate": 8.937065143526347e-05,
+      "loss": 0.2725,
       "step": 70
     },
     {
       "epoch": 2.84,
-      "grad_norm": 0.32281526923179626,
-      "learning_rate": 4.256713373170564e-05,
-      "loss": 0.343,
+      "grad_norm": 0.32543686032295227,
+      "learning_rate": 8.672744727162781e-05,
+      "loss": 0.2916,
       "step": 71
     },
     {
       "epoch": 2.88,
-      "grad_norm": 0.3217308223247528,
-      "learning_rate": 3.9891885271697496e-05,
-      "loss": 0.3142,
+      "grad_norm": 0.329887717962265,
+      "learning_rate": 8.409365039809281e-05,
+      "loss": 0.2883,
       "step": 72
     },
     {
       "epoch": 2.92,
-      "grad_norm": 0.3167681097984314,
-      "learning_rate": 3.7282364152646297e-05,
-      "loss": 0.3351,
+      "grad_norm": 0.33566269278526306,
+      "learning_rate": 8.147112759128859e-05,
+      "loss": 0.2912,
       "step": 73
     },
     {
       "epoch": 2.96,
-      "grad_norm": 0.3280259072780609,
-      "learning_rate": 3.4741423847583134e-05,
-      "loss": 0.316,
+      "grad_norm": 0.31171417236328125,
+      "learning_rate": 7.886173763703757e-05,
+      "loss": 0.2812,
       "step": 74
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.32448863983154297,
-      "learning_rate": 3.227184283742591e-05,
-      "loss": 0.3195,
+      "grad_norm": 0.31463706493377686,
+      "learning_rate": 7.626733001288851e-05,
+      "loss": 0.289,
       "step": 75
     },
     {
       "epoch": 3.04,
-      "grad_norm": 0.33512407541275024,
-      "learning_rate": 2.9876321572751144e-05,
-      "loss": 0.2814,
+      "grad_norm": 0.3288837671279907,
+      "learning_rate": 7.368974357724789e-05,
+      "loss": 0.2407,
       "step": 76
     },
     {
       "epoch": 3.08,
-      "grad_norm": 0.33603280782699585,
-      "learning_rate": 2.7557479520891104e-05,
-      "loss": 0.2981,
+      "grad_norm": 0.3028244376182556,
+      "learning_rate": 7.113080526603792e-05,
+      "loss": 0.214,
       "step": 77
     },
     {
       "epoch": 3.12,
-      "grad_norm": 0.33797961473464966,
-      "learning_rate": 2.5317852301584643e-05,
-      "loss": 0.293,
+      "grad_norm": 0.31008780002593994,
+      "learning_rate": 6.859232879780515e-05,
+      "loss": 0.2133,
       "step": 78
     },
     {
       "epoch": 3.16,
-      "grad_norm": 0.3338729739189148,
-      "learning_rate": 2.315988891431412e-05,
-      "loss": 0.2988,
+      "grad_norm": 0.32053515315055847,
+      "learning_rate": 6.607611338819697e-05,
+      "loss": 0.209,
       "step": 79
     },
     {
       "epoch": 3.2,
-      "grad_norm": 0.3124498128890991,
-      "learning_rate": 2.1085949060360654e-05,
-      "loss": 0.2956,
+      "grad_norm": 0.2950594425201416,
+      "learning_rate": 6.358394247471778e-05,
+      "loss": 0.213,
       "step": 80
     },
     {
       "epoch": 3.24,
-      "grad_norm": 0.30723869800567627,
-      "learning_rate": 1.9098300562505266e-05,
-      "loss": 0.2762,
+      "grad_norm": 0.30111971497535706,
+      "learning_rate": 6.111758245266794e-05,
+      "loss": 0.2218,
       "step": 81
     },
     {
       "epoch": 3.2800000000000002,
-      "grad_norm": 0.30877774953842163,
-      "learning_rate": 1.7199116885197995e-05,
-      "loss": 0.2811,
+      "grad_norm": 0.28454694151878357,
+      "learning_rate": 5.867878142316221e-05,
+      "loss": 0.2061,
       "step": 82
     },
     {
       "epoch": 3.32,
-      "grad_norm": 0.3040075898170471,
-      "learning_rate": 1.5390474757906446e-05,
-      "loss": 0.2843,
+      "grad_norm": 0.29927462339401245,
+      "learning_rate": 5.626926795411447e-05,
+      "loss": 0.2181,
       "step": 83
     },
     {
       "epoch": 3.36,
-      "grad_norm": 0.29099854826927185,
-      "learning_rate": 1.3674351904242611e-05,
-      "loss": 0.2823,
+      "grad_norm": 0.2855587899684906,
+      "learning_rate": 5.38907498550674e-05,
+      "loss": 0.2293,
       "step": 84
     },
     {
       "epoch": 3.4,
-      "grad_norm": 0.29945293068885803,
-      "learning_rate": 1.2052624879351104e-05,
-      "loss": 0.3007,
+      "grad_norm": 0.2874046862125397,
+      "learning_rate": 5.1544912966734994e-05,
+      "loss": 0.2089,
       "step": 85
     },
     {
       "epoch": 3.44,
-      "grad_norm": 0.29691964387893677,
-      "learning_rate": 1.0527067017923654e-05,
-      "loss": 0.2866,
+      "grad_norm": 0.3082650303840637,
+      "learning_rate": 4.9233419966116036e-05,
+      "loss": 0.213,
       "step": 86
     },
     {
       "epoch": 3.48,
-      "grad_norm": 0.29864245653152466,
-      "learning_rate": 9.09934649508375e-06,
-      "loss": 0.2946,
+      "grad_norm": 0.30149927735328674,
+      "learning_rate": 4.695790918802576e-05,
+      "loss": 0.2061,
       "step": 87
     },
     {
       "epoch": 3.52,
-      "grad_norm": 0.29509297013282776,
-      "learning_rate": 7.771024502261526e-06,
-      "loss": 0.2717,
+      "grad_norm": 0.32016074657440186,
+      "learning_rate": 4.47199934638807e-05,
+      "loss": 0.2284,
       "step": 88
     },
     {
       "epoch": 3.56,
-      "grad_norm": 0.3103945851325989,
-      "learning_rate": 6.543553540053926e-06,
-      "loss": 0.287,
+      "grad_norm": 0.30953240394592285,
+      "learning_rate": 4.252125897855932e-05,
+      "loss": 0.2231,
       "step": 89
     },
     {
       "epoch": 3.6,
-      "grad_norm": 0.2974102795124054,
-      "learning_rate": 5.418275829936537e-06,
-      "loss": 0.2815,
+      "grad_norm": 0.2966735363006592,
+      "learning_rate": 4.036326414614985e-05,
+      "loss": 0.2047,
       "step": 90
     },
     {
       "epoch": 3.64,
-      "grad_norm": 0.3041694760322571,
-      "learning_rate": 4.3964218465642355e-06,
-      "loss": 0.2905,
+      "grad_norm": 0.31812402606010437,
+      "learning_rate": 3.824753850538082e-05,
+      "loss": 0.2067,
       "step": 91
     },
     {
       "epoch": 3.68,
-      "grad_norm": 0.2940130829811096,
-      "learning_rate": 3.4791089722651436e-06,
-      "loss": 0.273,
+      "grad_norm": 0.3114762604236603,
+      "learning_rate": 3.617558163551802e-05,
+      "loss": 0.199,
       "step": 92
     },
     {
       "epoch": 3.7199999999999998,
-      "grad_norm": 0.2969824969768524,
-      "learning_rate": 2.667340275199426e-06,
-      "loss": 0.2691,
+      "grad_norm": 0.31614482402801514,
+      "learning_rate": 3.414886209349615e-05,
+      "loss": 0.2188,
       "step": 93
     },
     {
       "epoch": 3.76,
-      "grad_norm": 0.30998706817626953,
-      "learning_rate": 1.9620034125190644e-06,
-      "loss": 0.2984,
+      "grad_norm": 0.2987557053565979,
+      "learning_rate": 3.216881637303839e-05,
+      "loss": 0.1926,
       "step": 94
     },
     {
       "epoch": 3.8,
-      "grad_norm": 0.31238672137260437,
-      "learning_rate": 1.3638696597277679e-06,
-      "loss": 0.2856,
+      "grad_norm": 0.3159599304199219,
+      "learning_rate": 3.0236847886501542e-05,
+      "loss": 0.2143,
       "step": 95
     },
     {
       "epoch": 3.84,
-      "grad_norm": 0.29781845211982727,
-      "learning_rate": 8.735930673024806e-07,
-      "loss": 0.2911,
+      "grad_norm": 0.3066674470901489,
+      "learning_rate": 2.8354325970168484e-05,
+      "loss": 0.2035,
       "step": 96
     },
     {
       "epoch": 3.88,
-      "grad_norm": 0.29917433857917786,
-      "learning_rate": 4.917097454988584e-07,
-      "loss": 0.278,
+      "grad_norm": 0.3181123435497284,
+      "learning_rate": 2.6522584913693294e-05,
+      "loss": 0.1922,
       "step": 97
     },
     {
       "epoch": 3.92,
-      "grad_norm": 0.30273205041885376,
-      "learning_rate": 2.1863727812254653e-07,
-      "loss": 0.2811,
+      "grad_norm": 0.31559616327285767,
+      "learning_rate": 2.4742923014386156e-05,
+      "loss": 0.2322,
       "step": 98
     },
     {
       "epoch": 3.96,
-      "grad_norm": 0.3171875774860382,
-      "learning_rate": 5.467426590739511e-08,
-      "loss": 0.2876,
+      "grad_norm": 0.3333965539932251,
+      "learning_rate": 2.301660165700936e-05,
+      "loss": 0.2285,
       "step": 99
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.3021162450313568,
-      "learning_rate": 0.0,
-      "loss": 0.2926,
+      "grad_norm": 0.30800747871398926,
+      "learning_rate": 2.1344844419735755e-05,
+      "loss": 0.2196,
       "step": 100
+    },
+    {
+      "epoch": 4.04,
+      "grad_norm": 0.28937435150146484,
+      "learning_rate": 1.9728836206903656e-05,
+      "loss": 0.1743,
+      "step": 101
+    },
+    {
+      "epoch": 4.08,
+      "grad_norm": 0.2849326729774475,
+      "learning_rate": 1.8169722409183097e-05,
+      "loss": 0.1679,
+      "step": 102
+    },
+    {
+      "epoch": 4.12,
+      "grad_norm": 0.28659793734550476,
+      "learning_rate": 1.6668608091748495e-05,
+      "loss": 0.193,
+      "step": 103
+    },
+    {
+      "epoch": 4.16,
+      "grad_norm": 0.28456950187683105,
+      "learning_rate": 1.522655721103291e-05,
+      "loss": 0.1816,
+      "step": 104
+    },
+    {
+      "epoch": 4.2,
+      "grad_norm": 0.2843972444534302,
+      "learning_rate": 1.3844591860619383e-05,
+      "loss": 0.1712,
+      "step": 105
+    },
+    {
+      "epoch": 4.24,
+      "grad_norm": 0.28796839714050293,
+      "learning_rate": 1.2523691546803873e-05,
+      "loss": 0.1802,
+      "step": 106
+    },
+    {
+      "epoch": 4.28,
+      "grad_norm": 0.2841332256793976,
+      "learning_rate": 1.1264792494342857e-05,
+      "loss": 0.1833,
+      "step": 107
+    },
+    {
+      "epoch": 4.32,
+      "grad_norm": 0.2956565022468567,
+      "learning_rate": 1.0068786982878087e-05,
+      "loss": 0.1845,
+      "step": 108
+    },
+    {
+      "epoch": 4.36,
+      "grad_norm": 0.28327876329421997,
+      "learning_rate": 8.936522714508678e-06,
+      "loss": 0.1769,
+      "step": 109
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 0.2768532335758209,
+      "learning_rate": 7.868802212958703e-06,
+      "loss": 0.1745,
+      "step": 110
+    },
+    {
+      "epoch": 4.44,
+      "grad_norm": 0.27693644165992737,
+      "learning_rate": 6.866382254766157e-06,
+      "loss": 0.1818,
+      "step": 111
+    },
+    {
+      "epoch": 4.48,
+      "grad_norm": 0.2891138792037964,
+      "learning_rate": 5.929973332896677e-06,
+      "loss": 0.1858,
+      "step": 112
+    },
+    {
+      "epoch": 4.52,
+      "grad_norm": 0.2900177538394928,
+      "learning_rate": 5.060239153161872e-06,
+      "loss": 0.1811,
+      "step": 113
+    },
+    {
+      "epoch": 4.5600000000000005,
+      "grad_norm": 0.27316245436668396,
+      "learning_rate": 4.257796163799455e-06,
+      "loss": 0.1609,
+      "step": 114
+    },
+    {
+      "epoch": 4.6,
+      "grad_norm": 0.2771458923816681,
+      "learning_rate": 3.5232131185484076e-06,
+      "loss": 0.183,
+      "step": 115
+    },
+    {
+      "epoch": 4.64,
+      "grad_norm": 0.28434255719184875,
+      "learning_rate": 2.857010673529015e-06,
+      "loss": 0.1834,
+      "step": 116
+    },
+    {
+      "epoch": 4.68,
+      "grad_norm": 0.2891601026058197,
+      "learning_rate": 2.259661018213333e-06,
+      "loss": 0.1848,
+      "step": 117
+    },
+    {
+      "epoch": 4.72,
+      "grad_norm": 0.2810252010822296,
+      "learning_rate": 1.7315875407479032e-06,
+      "loss": 0.1866,
+      "step": 118
+    },
+    {
+      "epoch": 4.76,
+      "grad_norm": 0.28846290707588196,
+      "learning_rate": 1.2731645278655445e-06,
+      "loss": 0.1815,
+      "step": 119
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.2806852459907532,
+      "learning_rate": 8.847168995992916e-07,
+      "loss": 0.1865,
+      "step": 120
+    },
+    {
+      "epoch": 4.84,
+      "grad_norm": 0.27965104579925537,
+      "learning_rate": 5.665199789862907e-07,
+      "loss": 0.1803,
+      "step": 121
+    },
+    {
+      "epoch": 4.88,
+      "grad_norm": 0.2856048345565796,
+      "learning_rate": 3.1879929692498757e-07,
+      "loss": 0.1848,
+      "step": 122
+    },
+    {
+      "epoch": 4.92,
+      "grad_norm": 0.2768332064151764,
+      "learning_rate": 1.4173043232380557e-07,
+      "loss": 0.1866,
+      "step": 123
+    },
+    {
+      "epoch": 4.96,
+      "grad_norm": 0.2879635691642761,
+      "learning_rate": 3.5438887654737355e-08,
+      "loss": 0.1807,
+      "step": 124
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.2879068851470947,
+      "learning_rate": 0.0,
+      "loss": 0.1706,
+      "step": 125
     }
   ],
   "logging_steps": 1,
-  "max_steps": 100,
+  "max_steps": 125,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -726,7 +901,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.5000238750511923e+17,
+  "total_flos": 1.9085560601370624e+17,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07de7a446103efc6abf630ef72bd582d5c357a7607d0c9fdc45dd38d04c8d45d
+oid sha256:4261906a8e783b239ce01471e929840a74327be1581835e361bd20be86b6d3e2
 size 5240