Update fine-tuned model

adapter_config.json

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

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d6a977183e505b764d0dfd8b152287fb444c9cf075dabe509af0cdc23a3a374
+oid sha256:6b04265a272e2a30fc9bef6e6a0ef55963640a7d1960c9c2533161609f2a34a0
 size 159967880

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfa58f767b69164ca37825b684f8eff7d5bf21a6e04391029c45a229c5c64612
+oid sha256:d28d76aa254e618cba108a7f6830a033d2b732e6ba3b9fa8d00abf4650a8bb6f
 size 852876198

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8ad58ab3a10b5d635a4a58ef5dd2e25339fd79e7202a61585dca7b9458a71aa
+oid sha256:4a56085db13eeed8ac0f0c4df58b9a9441bede75240cec0a9b75bf446a43f781
 size 14244

trainer_state.json

@@ -10,877 +10,877 @@
   "log_history": [
     {
       "epoch": 0.04,
-      "grad_norm": 1.5257668495178223,
+      "grad_norm": 1.570319414138794,
       "learning_rate": 2.857142857142857e-05,
-      "loss": 0.9817,
+      "loss": 0.9956,
       "step": 1
     },
     {
       "epoch": 0.08,
-      "grad_norm": 1.4205083847045898,
+      "grad_norm": 1.5578975677490234,
       "learning_rate": 5.714285714285714e-05,
-      "loss": 0.9677,
+      "loss": 0.946,
       "step": 2
     },
     {
       "epoch": 0.12,
-      "grad_norm": 1.4500166177749634,
+      "grad_norm": 1.444091796875,
       "learning_rate": 8.571428571428571e-05,
-      "loss": 0.9229,
+      "loss": 0.9214,
       "step": 3
     },
     {
       "epoch": 0.16,
-      "grad_norm": 1.1406153440475464,
+      "grad_norm": 1.2263134717941284,
       "learning_rate": 0.00011428571428571428,
-      "loss": 0.8997,
+      "loss": 0.8926,
       "step": 4
     },
     {
       "epoch": 0.2,
-      "grad_norm": 0.9545271396636963,
+      "grad_norm": 0.912525475025177,
       "learning_rate": 0.00014285714285714287,
-      "loss": 0.8483,
+      "loss": 0.8399,
       "step": 5
     },
     {
       "epoch": 0.24,
-      "grad_norm": 0.8234877586364746,
+      "grad_norm": 0.7677531242370605,
       "learning_rate": 0.00017142857142857143,
-      "loss": 0.8351,
+      "loss": 0.7987,
       "step": 6
     },
     {
       "epoch": 0.28,
-      "grad_norm": 0.6187587380409241,
+      "grad_norm": 0.6140649914741516,
       "learning_rate": 0.0002,
-      "loss": 0.7575,
+      "loss": 0.7481,
       "step": 7
     },
     {
       "epoch": 0.32,
-      "grad_norm": 0.4950176775455475,
+      "grad_norm": 0.8730568885803223,
       "learning_rate": 0.00019996456111234527,
-      "loss": 0.7344,
+      "loss": 0.7544,
       "step": 8
     },
     {
       "epoch": 0.36,
-      "grad_norm": 0.43441540002822876,
+      "grad_norm": 0.7831456661224365,
       "learning_rate": 0.0001998582695676762,
-      "loss": 0.7035,
+      "loss": 0.7009,
       "step": 9
     },
     {
       "epoch": 0.4,
-      "grad_norm": 0.4230976402759552,
+      "grad_norm": 0.512580394744873,
       "learning_rate": 0.000199681200703075,
-      "loss": 0.7492,
+      "loss": 0.6704,
       "step": 10
     },
     {
       "epoch": 0.44,
-      "grad_norm": 0.3713449239730835,
+      "grad_norm": 0.37298211455345154,
       "learning_rate": 0.00019943348002101371,
-      "loss": 0.6831,
+      "loss": 0.7061,
       "step": 11
     },
     {
       "epoch": 0.48,
-      "grad_norm": 0.3475893437862396,
+      "grad_norm": 0.3507251739501953,
       "learning_rate": 0.00019911528310040074,
-      "loss": 0.7153,
+      "loss": 0.6529,
       "step": 12
     },
     {
       "epoch": 0.52,
-      "grad_norm": 0.3414737582206726,
+      "grad_norm": 0.3471398651599884,
       "learning_rate": 0.00019872683547213446,
-      "loss": 0.6943,
+      "loss": 0.6738,
       "step": 13
     },
     {
       "epoch": 0.56,
-      "grad_norm": 0.3290707767009735,
+      "grad_norm": 0.33570075035095215,
       "learning_rate": 0.00019826841245925212,
-      "loss": 0.6698,
+      "loss": 0.6441,
       "step": 14
     },
     {
       "epoch": 0.6,
-      "grad_norm": 0.3098256587982178,
+      "grad_norm": 0.31052011251449585,
       "learning_rate": 0.00019774033898178667,
-      "loss": 0.6385,
+      "loss": 0.6061,
       "step": 15
     },
     {
       "epoch": 0.64,
-      "grad_norm": 0.31097927689552307,
+      "grad_norm": 0.3227557837963104,
       "learning_rate": 0.00019714298932647098,
-      "loss": 0.6694,
+      "loss": 0.6389,
       "step": 16
     },
     {
       "epoch": 0.68,
-      "grad_norm": 0.32695478200912476,
+      "grad_norm": 0.32108747959136963,
       "learning_rate": 0.0001964767868814516,
-      "loss": 0.6702,
+      "loss": 0.6036,
       "step": 17
     },
     {
       "epoch": 0.72,
-      "grad_norm": 0.30651360750198364,
+      "grad_norm": 0.31963539123535156,
       "learning_rate": 0.00019574220383620055,
-      "loss": 0.6172,
+      "loss": 0.6126,
       "step": 18
     },
     {
       "epoch": 0.76,
-      "grad_norm": 0.31658628582954407,
+      "grad_norm": 0.3213265538215637,
       "learning_rate": 0.00019493976084683813,
-      "loss": 0.6248,
+      "loss": 0.6219,
       "step": 19
     },
     {
       "epoch": 0.8,
-      "grad_norm": 0.30691343545913696,
+      "grad_norm": 0.32397329807281494,
       "learning_rate": 0.00019407002666710336,
-      "loss": 0.6281,
+      "loss": 0.6241,
       "step": 20
     },
     {
       "epoch": 0.84,
-      "grad_norm": 0.29928600788116455,
+      "grad_norm": 0.31370866298675537,
       "learning_rate": 0.00019313361774523385,
-      "loss": 0.6208,
+      "loss": 0.5877,
       "step": 21
     },
     {
       "epoch": 0.88,
-      "grad_norm": 0.30110830068588257,
+      "grad_norm": 0.3048774302005768,
       "learning_rate": 0.00019213119778704128,
-      "loss": 0.5995,
+      "loss": 0.5939,
       "step": 22
     },
     {
       "epoch": 0.92,
-      "grad_norm": 0.2957627475261688,
+      "grad_norm": 0.31822916865348816,
       "learning_rate": 0.00019106347728549135,
-      "loss": 0.6016,
+      "loss": 0.5762,
       "step": 23
     },
     {
       "epoch": 0.96,
-      "grad_norm": 0.29468321800231934,
+      "grad_norm": 0.309684693813324,
       "learning_rate": 0.00018993121301712193,
-      "loss": 0.5997,
+      "loss": 0.5746,
       "step": 24
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.2975432276725769,
+      "grad_norm": 0.3195765018463135,
       "learning_rate": 0.00018873520750565718,
-      "loss": 0.6106,
+      "loss": 0.5715,
       "step": 25
     },
     {
       "epoch": 1.04,
-      "grad_norm": 0.31368955969810486,
+      "grad_norm": 0.3123801052570343,
       "learning_rate": 0.00018747630845319612,
-      "loss": 0.4904,
+      "loss": 0.4747,
       "step": 26
     },
     {
       "epoch": 1.08,
-      "grad_norm": 0.31520742177963257,
+      "grad_norm": 0.31075745820999146,
       "learning_rate": 0.0001861554081393806,
-      "loss": 0.5047,
+      "loss": 0.4722,
       "step": 27
     },
     {
       "epoch": 1.12,
-      "grad_norm": 0.3065387010574341,
+      "grad_norm": 0.31425103545188904,
       "learning_rate": 0.0001847734427889671,
-      "loss": 0.4789,
+      "loss": 0.4374,
       "step": 28
     },
     {
       "epoch": 1.16,
-      "grad_norm": 0.2928008437156677,
+      "grad_norm": 0.28231778740882874,
       "learning_rate": 0.0001833313919082515,
-      "loss": 0.4898,
+      "loss": 0.4373,
       "step": 29
     },
     {
       "epoch": 1.2,
-      "grad_norm": 0.2972809672355652,
+      "grad_norm": 0.30015143752098083,
       "learning_rate": 0.0001818302775908169,
-      "loss": 0.4904,
+      "loss": 0.4551,
       "step": 30
     },
     {
       "epoch": 1.24,
-      "grad_norm": 0.2960495948791504,
+      "grad_norm": 0.28502359986305237,
       "learning_rate": 0.00018027116379309638,
-      "loss": 0.4629,
+      "loss": 0.4322,
       "step": 31
     },
     {
       "epoch": 1.28,
-      "grad_norm": 0.28314685821533203,
+      "grad_norm": 0.28252413868904114,
       "learning_rate": 0.00017865515558026428,
-      "loss": 0.4483,
+      "loss": 0.4234,
       "step": 32
     },
     {
       "epoch": 1.32,
-      "grad_norm": 0.2923860549926758,
+      "grad_norm": 0.30743300914764404,
       "learning_rate": 0.00017698339834299061,
-      "loss": 0.4602,
+      "loss": 0.444,
       "step": 33
     },
     {
       "epoch": 1.3599999999999999,
-      "grad_norm": 0.2865493893623352,
+      "grad_norm": 0.2907680869102478,
       "learning_rate": 0.00017525707698561385,
-      "loss": 0.4565,
+      "loss": 0.4154,
       "step": 34
     },
     {
       "epoch": 1.4,
-      "grad_norm": 0.29644760489463806,
+      "grad_norm": 0.30981379747390747,
       "learning_rate": 0.00017347741508630672,
-      "loss": 0.4491,
+      "loss": 0.4154,
       "step": 35
     },
     {
       "epoch": 1.44,
-      "grad_norm": 0.2860000729560852,
+      "grad_norm": 0.2963266968727112,
       "learning_rate": 0.00017164567402983152,
-      "loss": 0.444,
+      "loss": 0.404,
       "step": 36
     },
     {
       "epoch": 1.48,
-      "grad_norm": 0.2956458330154419,
+      "grad_norm": 0.3006102740764618,
       "learning_rate": 0.0001697631521134985,
-      "loss": 0.464,
+      "loss": 0.4141,
       "step": 37
     },
     {
       "epoch": 1.52,
-      "grad_norm": 0.30596640706062317,
+      "grad_norm": 0.3143654763698578,
       "learning_rate": 0.00016783118362696163,
-      "loss": 0.4596,
+      "loss": 0.3708,
       "step": 38
     },
     {
       "epoch": 1.56,
-      "grad_norm": 0.31658679246902466,
+      "grad_norm": 0.3136065602302551,
       "learning_rate": 0.00016585113790650388,
-      "loss": 0.4184,
+      "loss": 0.4031,
       "step": 39
     },
     {
       "epoch": 1.6,
-      "grad_norm": 0.3067062497138977,
+      "grad_norm": 0.31369641423225403,
       "learning_rate": 0.00016382441836448202,
-      "loss": 0.3952,
+      "loss": 0.3937,
       "step": 40
     },
     {
       "epoch": 1.6400000000000001,
-      "grad_norm": 0.30973076820373535,
+      "grad_norm": 0.3261353671550751,
       "learning_rate": 0.0001617524614946192,
-      "loss": 0.4109,
+      "loss": 0.3931,
       "step": 41
     },
     {
       "epoch": 1.6800000000000002,
-      "grad_norm": 0.30597272515296936,
+      "grad_norm": 0.3178551197052002,
       "learning_rate": 0.00015963673585385016,
-      "loss": 0.3853,
+      "loss": 0.3829,
       "step": 42
     },
     {
       "epoch": 1.72,
-      "grad_norm": 0.3238050639629364,
+      "grad_norm": 0.3346520960330963,
       "learning_rate": 0.0001574787410214407,
-      "loss": 0.4074,
+      "loss": 0.3792,
       "step": 43
     },
     {
       "epoch": 1.76,
-      "grad_norm": 0.32239633798599243,
+      "grad_norm": 0.3492076098918915,
       "learning_rate": 0.00015528000653611935,
-      "loss": 0.421,
+      "loss": 0.3777,
       "step": 44
     },
     {
       "epoch": 1.8,
-      "grad_norm": 0.322157084941864,
+      "grad_norm": 0.3226877748966217,
       "learning_rate": 0.00015304209081197425,
-      "loss": 0.4156,
+      "loss": 0.379,
       "step": 45
     },
     {
       "epoch": 1.8399999999999999,
-      "grad_norm": 0.32195115089416504,
+      "grad_norm": 0.31526848673820496,
       "learning_rate": 0.000150766580033884,
-      "loss": 0.3955,
+      "loss": 0.3848,
       "step": 46
     },
     {
       "epoch": 1.88,
-      "grad_norm": 0.312031626701355,
+      "grad_norm": 0.3140419125556946,
       "learning_rate": 0.00014845508703326504,
-      "loss": 0.3814,
+      "loss": 0.3542,
       "step": 47
     },
     {
       "epoch": 1.92,
-      "grad_norm": 0.31998252868652344,
+      "grad_norm": 0.32493019104003906,
       "learning_rate": 0.0001461092501449326,
-      "loss": 0.4039,
+      "loss": 0.3842,
       "step": 48
     },
     {
       "epoch": 1.96,
-      "grad_norm": 0.32456570863723755,
+      "grad_norm": 0.33945709466934204,
       "learning_rate": 0.00014373073204588556,
-      "loss": 0.3963,
+      "loss": 0.3677,
       "step": 49
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.32449814677238464,
+      "grad_norm": 0.3231065571308136,
       "learning_rate": 0.00014132121857683783,
-      "loss": 0.3845,
+      "loss": 0.3752,
       "step": 50
     },
     {
       "epoch": 2.04,
-      "grad_norm": 0.31362128257751465,
+      "grad_norm": 0.2973474860191345,
       "learning_rate": 0.00013888241754733208,
-      "loss": 0.3138,
+      "loss": 0.2789,
       "step": 51
     },
     {
       "epoch": 2.08,
-      "grad_norm": 0.30638328194618225,
+      "grad_norm": 0.30944961309432983,
       "learning_rate": 0.00013641605752528224,
-      "loss": 0.317,
+      "loss": 0.279,
       "step": 52
     },
     {
       "epoch": 2.12,
-      "grad_norm": 0.32126715779304504,
+      "grad_norm": 0.3238925337791443,
       "learning_rate": 0.00013392388661180303,
-      "loss": 0.3004,
+      "loss": 0.2919,
       "step": 53
     },
     {
       "epoch": 2.16,
-      "grad_norm": 0.28879204392433167,
+      "grad_norm": 0.30867430567741394,
       "learning_rate": 0.0001314076712021949,
-      "loss": 0.2997,
+      "loss": 0.2998,
       "step": 54
     },
     {
       "epoch": 2.2,
-      "grad_norm": 0.29814431071281433,
+      "grad_norm": 0.29136329889297485,
       "learning_rate": 0.0001288691947339621,
-      "loss": 0.2934,
+      "loss": 0.279,
       "step": 55
     },
     {
       "epoch": 2.24,
-      "grad_norm": 0.28737854957580566,
+      "grad_norm": 0.2846209704875946,
       "learning_rate": 0.00012631025642275212,
-      "loss": 0.3185,
+      "loss": 0.2826,
       "step": 56
     },
     {
       "epoch": 2.2800000000000002,
-      "grad_norm": 0.30604204535484314,
+      "grad_norm": 0.29411131143569946,
       "learning_rate": 0.0001237326699871115,
-      "loss": 0.3059,
+      "loss": 0.2693,
       "step": 57
     },
     {
       "epoch": 2.32,
-      "grad_norm": 0.2846826910972595,
+      "grad_norm": 0.2946978211402893,
       "learning_rate": 0.00012113826236296244,
-      "loss": 0.2779,
+      "loss": 0.2777,
       "step": 58
     },
     {
       "epoch": 2.36,
-      "grad_norm": 0.3072626292705536,
+      "grad_norm": 0.30307185649871826,
       "learning_rate": 0.00011852887240871145,
-      "loss": 0.2958,
+      "loss": 0.2673,
       "step": 59
     },
     {
       "epoch": 2.4,
-      "grad_norm": 0.30152225494384766,
+      "grad_norm": 0.31018176674842834,
       "learning_rate": 0.00011590634960190721,
-      "loss": 0.276,
+      "loss": 0.2658,
       "step": 60
     },
     {
       "epoch": 2.44,
-      "grad_norm": 0.31316888332366943,
+      "grad_norm": 0.30534908175468445,
       "learning_rate": 0.00011327255272837221,
-      "loss": 0.2863,
+      "loss": 0.2669,
       "step": 61
     },
     {
       "epoch": 2.48,
-      "grad_norm": 0.33458268642425537,
+      "grad_norm": 0.3161245286464691,
       "learning_rate": 0.00011062934856473655,
-      "loss": 0.2999,
+      "loss": 0.2793,
       "step": 62
     },
     {
       "epoch": 2.52,
-      "grad_norm": 0.3459411561489105,
+      "grad_norm": 0.3256876766681671,
       "learning_rate": 0.00010797861055530831,
-      "loss": 0.2861,
+      "loss": 0.2904,
       "step": 63
     },
     {
       "epoch": 2.56,
-      "grad_norm": 0.34091514348983765,
+      "grad_norm": 0.32278570532798767,
       "learning_rate": 0.00010532221748421787,
-      "loss": 0.2697,
+      "loss": 0.2785,
       "step": 64
     },
     {
       "epoch": 2.6,
-      "grad_norm": 0.3289962708950043,
+      "grad_norm": 0.3227723240852356,
       "learning_rate": 0.00010266205214377748,
-      "loss": 0.2722,
+      "loss": 0.2655,
       "step": 65
     },
     {
       "epoch": 2.64,
-      "grad_norm": 0.324381023645401,
+      "grad_norm": 0.336511492729187,
       "learning_rate": 0.0001,
-      "loss": 0.2864,
+      "loss": 0.264,
       "step": 66
     },
     {
       "epoch": 2.68,
-      "grad_norm": 0.34481576085090637,
+      "grad_norm": 0.33010682463645935,
       "learning_rate": 9.733794785622253e-05,
-      "loss": 0.3002,
+      "loss": 0.2705,
       "step": 67
     },
     {
       "epoch": 2.7199999999999998,
-      "grad_norm": 0.33097216486930847,
+      "grad_norm": 0.33406832814216614,
       "learning_rate": 9.467778251578217e-05,
-      "loss": 0.2915,
+      "loss": 0.2587,
       "step": 68
     },
     {
       "epoch": 2.76,
-      "grad_norm": 0.3260320723056793,
+      "grad_norm": 0.31980621814727783,
       "learning_rate": 9.202138944469168e-05,
-      "loss": 0.2962,
+      "loss": 0.2554,
       "step": 69
     },
     {
       "epoch": 2.8,
-      "grad_norm": 0.3088409900665283,
+      "grad_norm": 0.3083789348602295,
       "learning_rate": 8.937065143526347e-05,
-      "loss": 0.2725,
+      "loss": 0.2594,
       "step": 70
     },
     {
       "epoch": 2.84,
-      "grad_norm": 0.32543686032295227,
+      "grad_norm": 0.34898000955581665,
       "learning_rate": 8.672744727162781e-05,
-      "loss": 0.2916,
+      "loss": 0.2741,
       "step": 71
     },
     {
       "epoch": 2.88,
-      "grad_norm": 0.329887717962265,
+      "grad_norm": 0.3146231174468994,
       "learning_rate": 8.409365039809281e-05,
-      "loss": 0.2883,
+      "loss": 0.2565,
       "step": 72
     },
     {
       "epoch": 2.92,
-      "grad_norm": 0.33566269278526306,
+      "grad_norm": 0.3152382969856262,
       "learning_rate": 8.147112759128859e-05,
-      "loss": 0.2912,
+      "loss": 0.2492,
       "step": 73
     },
     {
       "epoch": 2.96,
-      "grad_norm": 0.31171417236328125,
+      "grad_norm": 0.32025042176246643,
       "learning_rate": 7.886173763703757e-05,
-      "loss": 0.2812,
+      "loss": 0.2589,
       "step": 74
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.31463706493377686,
+      "grad_norm": 0.3296203911304474,
       "learning_rate": 7.626733001288851e-05,
-      "loss": 0.289,
+      "loss": 0.2622,
       "step": 75
     },
     {
       "epoch": 3.04,
-      "grad_norm": 0.3288837671279907,
+      "grad_norm": 0.29241126775741577,
       "learning_rate": 7.368974357724789e-05,
-      "loss": 0.2407,
+      "loss": 0.2021,
       "step": 76
     },
     {
       "epoch": 3.08,
-      "grad_norm": 0.3028244376182556,
+      "grad_norm": 0.2925822138786316,
       "learning_rate": 7.113080526603792e-05,
-      "loss": 0.214,
+      "loss": 0.215,
       "step": 77
     },
     {
       "epoch": 3.12,
-      "grad_norm": 0.31008780002593994,
+      "grad_norm": 0.29066669940948486,
       "learning_rate": 6.859232879780515e-05,
-      "loss": 0.2133,
+      "loss": 0.2078,
       "step": 78
     },
     {
       "epoch": 3.16,
-      "grad_norm": 0.32053515315055847,
+      "grad_norm": 0.3053090274333954,
       "learning_rate": 6.607611338819697e-05,
-      "loss": 0.209,
+      "loss": 0.2112,
       "step": 79
     },
     {
       "epoch": 3.2,
-      "grad_norm": 0.2950594425201416,
+      "grad_norm": 0.27980366349220276,
       "learning_rate": 6.358394247471778e-05,
-      "loss": 0.213,
+      "loss": 0.1845,
       "step": 80
     },
     {
       "epoch": 3.24,
-      "grad_norm": 0.30111971497535706,
+      "grad_norm": 0.2979828715324402,
       "learning_rate": 6.111758245266794e-05,
-      "loss": 0.2218,
+      "loss": 0.2063,
       "step": 81
     },
     {
       "epoch": 3.2800000000000002,
-      "grad_norm": 0.28454694151878357,
+      "grad_norm": 0.3061892092227936,
       "learning_rate": 5.867878142316221e-05,
-      "loss": 0.2061,
+      "loss": 0.2004,
       "step": 82
     },
     {
       "epoch": 3.32,
-      "grad_norm": 0.29927462339401245,
+      "grad_norm": 0.2933078408241272,
       "learning_rate": 5.626926795411447e-05,
-      "loss": 0.2181,
+      "loss": 0.2112,
       "step": 83
     },
     {
       "epoch": 3.36,
-      "grad_norm": 0.2855587899684906,
+      "grad_norm": 0.27492034435272217,
       "learning_rate": 5.38907498550674e-05,
-      "loss": 0.2293,
+      "loss": 0.1915,
       "step": 84
     },
     {
       "epoch": 3.4,
-      "grad_norm": 0.2874046862125397,
+      "grad_norm": 0.28737884759902954,
       "learning_rate": 5.1544912966734994e-05,
-      "loss": 0.2089,
+      "loss": 0.2102,
       "step": 85
     },
     {
       "epoch": 3.44,
-      "grad_norm": 0.3082650303840637,
+      "grad_norm": 0.3012617528438568,
       "learning_rate": 4.9233419966116036e-05,
-      "loss": 0.213,
+      "loss": 0.2003,
       "step": 86
     },
     {
       "epoch": 3.48,
-      "grad_norm": 0.30149927735328674,
+      "grad_norm": 0.2871868908405304,
       "learning_rate": 4.695790918802576e-05,
-      "loss": 0.2061,
+      "loss": 0.1934,
       "step": 87
     },
     {
       "epoch": 3.52,
-      "grad_norm": 0.32016074657440186,
+      "grad_norm": 0.29408469796180725,
       "learning_rate": 4.47199934638807e-05,
-      "loss": 0.2284,
+      "loss": 0.2067,
       "step": 88
     },
     {
       "epoch": 3.56,
-      "grad_norm": 0.30953240394592285,
+      "grad_norm": 0.27776145935058594,
       "learning_rate": 4.252125897855932e-05,
-      "loss": 0.2231,
+      "loss": 0.194,
       "step": 89
     },
     {
       "epoch": 3.6,
-      "grad_norm": 0.2966735363006592,
+      "grad_norm": 0.2886578440666199,
       "learning_rate": 4.036326414614985e-05,
-      "loss": 0.2047,
+      "loss": 0.1914,
       "step": 90
     },
     {
       "epoch": 3.64,
-      "grad_norm": 0.31812402606010437,
+      "grad_norm": 0.3070749342441559,
       "learning_rate": 3.824753850538082e-05,
-      "loss": 0.2067,
+      "loss": 0.2119,
       "step": 91
     },
     {
       "epoch": 3.68,
-      "grad_norm": 0.3114762604236603,
+      "grad_norm": 0.30975136160850525,
       "learning_rate": 3.617558163551802e-05,
-      "loss": 0.199,
+      "loss": 0.1958,
       "step": 92
     },
     {
       "epoch": 3.7199999999999998,
-      "grad_norm": 0.31614482402801514,
+      "grad_norm": 0.31414180994033813,
       "learning_rate": 3.414886209349615e-05,
-      "loss": 0.2188,
+      "loss": 0.1937,
       "step": 93
     },
     {
       "epoch": 3.76,
-      "grad_norm": 0.2987557053565979,
+      "grad_norm": 0.31353330612182617,
       "learning_rate": 3.216881637303839e-05,
-      "loss": 0.1926,
+      "loss": 0.1946,
       "step": 94
     },
     {
       "epoch": 3.8,
-      "grad_norm": 0.3159599304199219,
+      "grad_norm": 0.29970917105674744,
       "learning_rate": 3.0236847886501542e-05,
-      "loss": 0.2143,
+      "loss": 0.2061,
       "step": 95
     },
     {
       "epoch": 3.84,
-      "grad_norm": 0.3066674470901489,
+      "grad_norm": 0.30475103855133057,
       "learning_rate": 2.8354325970168484e-05,
-      "loss": 0.2035,
+      "loss": 0.1884,
       "step": 96
     },
     {
       "epoch": 3.88,
-      "grad_norm": 0.3181123435497284,
+      "grad_norm": 0.2989894449710846,
       "learning_rate": 2.6522584913693294e-05,
-      "loss": 0.1922,
+      "loss": 0.1924,
       "step": 97
     },
     {
       "epoch": 3.92,
-      "grad_norm": 0.31559616327285767,
+      "grad_norm": 0.30253222584724426,
       "learning_rate": 2.4742923014386156e-05,
-      "loss": 0.2322,
+      "loss": 0.1928,
       "step": 98
     },
     {
       "epoch": 3.96,
-      "grad_norm": 0.3333965539932251,
+      "grad_norm": 0.30798980593681335,
       "learning_rate": 2.301660165700936e-05,
-      "loss": 0.2285,
+      "loss": 0.2003,
       "step": 99
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.30800747871398926,
+      "grad_norm": 0.29617148637771606,
       "learning_rate": 2.1344844419735755e-05,
-      "loss": 0.2196,
+      "loss": 0.1867,
       "step": 100
     },
     {
       "epoch": 4.04,
-      "grad_norm": 0.28937435150146484,
+      "grad_norm": 0.2711721360683441,
       "learning_rate": 1.9728836206903656e-05,
-      "loss": 0.1743,
+      "loss": 0.1657,
       "step": 101
     },
     {
       "epoch": 4.08,
-      "grad_norm": 0.2849326729774475,
+      "grad_norm": 0.2831830382347107,
       "learning_rate": 1.8169722409183097e-05,
-      "loss": 0.1679,
+      "loss": 0.1692,
       "step": 102
     },
     {
       "epoch": 4.12,
-      "grad_norm": 0.28659793734550476,
+      "grad_norm": 0.272053599357605,
       "learning_rate": 1.6668608091748495e-05,
-      "loss": 0.193,
+      "loss": 0.1679,
       "step": 103
     },
     {
       "epoch": 4.16,
-      "grad_norm": 0.28456950187683105,
+      "grad_norm": 0.27920499444007874,
       "learning_rate": 1.522655721103291e-05,
-      "loss": 0.1816,
+      "loss": 0.1843,
       "step": 104
     },
     {
       "epoch": 4.2,
-      "grad_norm": 0.2843972444534302,
+      "grad_norm": 0.2759131193161011,
       "learning_rate": 1.3844591860619383e-05,
-      "loss": 0.1712,
+      "loss": 0.1626,
       "step": 105
     },
     {
       "epoch": 4.24,
-      "grad_norm": 0.28796839714050293,
+      "grad_norm": 0.2894749343395233,
       "learning_rate": 1.2523691546803873e-05,
-      "loss": 0.1802,
+      "loss": 0.1762,
       "step": 106
     },
     {
       "epoch": 4.28,
-      "grad_norm": 0.2841332256793976,
+      "grad_norm": 0.2730141580104828,
       "learning_rate": 1.1264792494342857e-05,
-      "loss": 0.1833,
+      "loss": 0.1703,
       "step": 107
     },
     {
       "epoch": 4.32,
-      "grad_norm": 0.2956565022468567,
+      "grad_norm": 0.26933005452156067,
       "learning_rate": 1.0068786982878087e-05,
-      "loss": 0.1845,
+      "loss": 0.163,
       "step": 108
     },
     {
       "epoch": 4.36,
-      "grad_norm": 0.28327876329421997,
+      "grad_norm": 0.2708563506603241,
       "learning_rate": 8.936522714508678e-06,
-      "loss": 0.1769,
+      "loss": 0.1679,
       "step": 109
     },
     {
       "epoch": 4.4,
-      "grad_norm": 0.2768532335758209,
+      "grad_norm": 0.26494595408439636,
       "learning_rate": 7.868802212958703e-06,
-      "loss": 0.1745,
+      "loss": 0.1611,
       "step": 110
     },
     {
       "epoch": 4.44,
-      "grad_norm": 0.27693644165992737,
+      "grad_norm": 0.2679324746131897,
       "learning_rate": 6.866382254766157e-06,
-      "loss": 0.1818,
+      "loss": 0.1646,
       "step": 111
     },
     {
       "epoch": 4.48,
-      "grad_norm": 0.2891138792037964,
+      "grad_norm": 0.26407480239868164,
       "learning_rate": 5.929973332896677e-06,
-      "loss": 0.1858,
+      "loss": 0.1616,
       "step": 112
     },
     {
       "epoch": 4.52,
-      "grad_norm": 0.2900177538394928,
+      "grad_norm": 0.28398773074150085,
       "learning_rate": 5.060239153161872e-06,
-      "loss": 0.1811,
+      "loss": 0.1795,
       "step": 113
     },
     {
       "epoch": 4.5600000000000005,
-      "grad_norm": 0.27316245436668396,
+      "grad_norm": 0.26915472745895386,
       "learning_rate": 4.257796163799455e-06,
-      "loss": 0.1609,
+      "loss": 0.1616,
       "step": 114
     },
     {
       "epoch": 4.6,
-      "grad_norm": 0.2771458923816681,
+      "grad_norm": 0.2760821282863617,
       "learning_rate": 3.5232131185484076e-06,
-      "loss": 0.183,
+      "loss": 0.1535,
       "step": 115
     },
     {
       "epoch": 4.64,
-      "grad_norm": 0.28434255719184875,
+      "grad_norm": 0.2624204754829407,
       "learning_rate": 2.857010673529015e-06,
-      "loss": 0.1834,
+      "loss": 0.1697,
       "step": 116
     },
     {
       "epoch": 4.68,
-      "grad_norm": 0.2891601026058197,
+      "grad_norm": 0.2621574103832245,
       "learning_rate": 2.259661018213333e-06,
-      "loss": 0.1848,
+      "loss": 0.1627,
       "step": 117
     },
     {
       "epoch": 4.72,
-      "grad_norm": 0.2810252010822296,
+      "grad_norm": 0.2760251760482788,
       "learning_rate": 1.7315875407479032e-06,
-      "loss": 0.1866,
+      "loss": 0.1747,
       "step": 118
     },
     {
       "epoch": 4.76,
-      "grad_norm": 0.28846290707588196,
+      "grad_norm": 0.26422515511512756,
       "learning_rate": 1.2731645278655445e-06,
-      "loss": 0.1815,
+      "loss": 0.1653,
       "step": 119
     },
     {
       "epoch": 4.8,
-      "grad_norm": 0.2806852459907532,
+      "grad_norm": 0.2870921790599823,
       "learning_rate": 8.847168995992916e-07,
-      "loss": 0.1865,
+      "loss": 0.1699,
       "step": 120
     },
     {
       "epoch": 4.84,
-      "grad_norm": 0.27965104579925537,
+      "grad_norm": 0.2736106812953949,
       "learning_rate": 5.665199789862907e-07,
-      "loss": 0.1803,
+      "loss": 0.1688,
       "step": 121
     },
     {
       "epoch": 4.88,
-      "grad_norm": 0.2856048345565796,
+      "grad_norm": 0.27442148327827454,
       "learning_rate": 3.1879929692498757e-07,
-      "loss": 0.1848,
+      "loss": 0.1722,
       "step": 122
     },
     {
       "epoch": 4.92,
-      "grad_norm": 0.2768332064151764,
+      "grad_norm": 0.26734665036201477,
       "learning_rate": 1.4173043232380557e-07,
-      "loss": 0.1866,
+      "loss": 0.1634,
       "step": 123
     },
     {
       "epoch": 4.96,
-      "grad_norm": 0.2879635691642761,
+      "grad_norm": 0.263090580701828,
       "learning_rate": 3.5438887654737355e-08,
-      "loss": 0.1807,
+      "loss": 0.1631,
       "step": 124
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.2879068851470947,
+      "grad_norm": 0.2877998650074005,
       "learning_rate": 0.0,
-      "loss": 0.1706,
+      "loss": 0.1683,
       "step": 125
     }
   ],
@@ -901,7 +901,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.9085560601370624e+17,
+  "total_flos": 1.8525301616443392e+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 sha256:4261906a8e783b239ce01471e929840a74327be1581835e361bd20be86b6d3e2
+oid sha256:70a4db1990766c1131233cf57ed77276a62f7f596ba0f44f8796b361645f9030
 size 5240