Update fine-tuned model

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "o_proj",
+    "up_proj",
+    "q_proj",
+    "gate_proj",
     "down_proj",
     "k_proj",
-    "gate_proj",
-    "v_proj",
-    "q_proj",
-    "up_proj",
-    "o_proj"
+    "v_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:c7590911661892e3c9392347ebda1de5380f253ffac55e85329e8a73e50b15cf
+oid sha256:6f6a49f69351456485e9eee6012dfa1780295775f8d7d1ac3267aded02fb47ea
 size 159967880

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36f202410c6d81593d68b1f21d20796871070444fabd294bacb88f27ee680fbe
+oid sha256:06728820a0d2910a7c148685fd6ca4a6f0ed059643904223033daa0932396a53
 size 852876198

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:809f3a6e13b93e710ea3d5ab2176d5175aca70d252bdab23bcb3fb5dd85b6047
+oid sha256:d02782d602efcdfe0b1c7a56f8520f6d57aed1961efb8e2e93c91f029291f316
 size 14244

scheduler.pt

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

trainer_state.json

@@ -1,893 +1,718 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 5.0,
+  "epoch": 4.0,
   "eval_steps": 500,
-  "global_step": 125,
+  "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.04,
-      "grad_norm": 0.5675837397575378,
-      "learning_rate": 2.857142857142857e-05,
-      "loss": 0.6417,
+      "grad_norm": 2.0647521018981934,
+      "learning_rate": 4e-05,
+      "loss": 1.1825,
       "step": 1
     },
     {
       "epoch": 0.08,
-      "grad_norm": 0.5383580923080444,
-      "learning_rate": 5.714285714285714e-05,
-      "loss": 0.6334,
+      "grad_norm": 1.9963816404342651,
+      "learning_rate": 8e-05,
+      "loss": 1.1448,
       "step": 2
     },
     {
       "epoch": 0.12,
-      "grad_norm": 0.523246705532074,
-      "learning_rate": 8.571428571428571e-05,
-      "loss": 0.6413,
+      "grad_norm": 1.82085120677948,
+      "learning_rate": 0.00012,
+      "loss": 1.1246,
       "step": 3
     },
     {
       "epoch": 0.16,
-      "grad_norm": 0.4905203580856323,
-      "learning_rate": 0.00011428571428571428,
-      "loss": 0.665,
+      "grad_norm": 1.3797991275787354,
+      "learning_rate": 0.00016,
+      "loss": 0.9428,
       "step": 4
     },
     {
       "epoch": 0.2,
-      "grad_norm": 0.5177513957023621,
-      "learning_rate": 0.00014285714285714287,
-      "loss": 0.6041,
+      "grad_norm": 1.0062527656555176,
+      "learning_rate": 0.0002,
+      "loss": 0.8782,
       "step": 5
     },
     {
       "epoch": 0.24,
-      "grad_norm": 0.482448011636734,
-      "learning_rate": 0.00017142857142857143,
-      "loss": 0.6078,
+      "grad_norm": 0.6854935884475708,
+      "learning_rate": 0.00019994532573409262,
+      "loss": 0.8182,
       "step": 6
     },
     {
       "epoch": 0.28,
-      "grad_norm": 0.49536892771720886,
-      "learning_rate": 0.0002,
-      "loss": 0.6412,
+      "grad_norm": 0.6088404059410095,
+      "learning_rate": 0.00019978136272187747,
+      "loss": 0.7337,
       "step": 7
     },
     {
       "epoch": 0.32,
-      "grad_norm": 0.5152004361152649,
-      "learning_rate": 0.00019996456111234527,
-      "loss": 0.607,
+      "grad_norm": 0.5058557391166687,
+      "learning_rate": 0.00019950829025450114,
+      "loss": 0.7606,
       "step": 8
     },
     {
       "epoch": 0.36,
-      "grad_norm": 0.496277391910553,
-      "learning_rate": 0.0001998582695676762,
-      "loss": 0.6227,
+      "grad_norm": 0.4383871853351593,
+      "learning_rate": 0.00019912640693269752,
+      "loss": 0.7934,
       "step": 9
     },
     {
       "epoch": 0.4,
-      "grad_norm": 0.5162248611450195,
-      "learning_rate": 0.000199681200703075,
-      "loss": 0.6012,
+      "grad_norm": 0.3763003349304199,
+      "learning_rate": 0.00019863613034027224,
+      "loss": 0.7623,
       "step": 10
     },
     {
       "epoch": 0.44,
-      "grad_norm": 0.5029735565185547,
-      "learning_rate": 0.00019943348002101371,
-      "loss": 0.5842,
+      "grad_norm": 0.3510531187057495,
+      "learning_rate": 0.00019803799658748094,
+      "loss": 0.7043,
       "step": 11
     },
     {
       "epoch": 0.48,
-      "grad_norm": 0.5039064884185791,
-      "learning_rate": 0.00019911528310040074,
-      "loss": 0.5833,
+      "grad_norm": 0.3534976541996002,
+      "learning_rate": 0.0001973326597248006,
+      "loss": 0.7702,
       "step": 12
     },
     {
       "epoch": 0.52,
-      "grad_norm": 0.486408531665802,
-      "learning_rate": 0.00019872683547213446,
-      "loss": 0.57,
+      "grad_norm": 0.33237403631210327,
+      "learning_rate": 0.00019652089102773488,
+      "loss": 0.7074,
       "step": 13
     },
     {
       "epoch": 0.56,
-      "grad_norm": 0.48064252734184265,
-      "learning_rate": 0.00019826841245925212,
-      "loss": 0.5857,
+      "grad_norm": 0.3145230710506439,
+      "learning_rate": 0.00019560357815343577,
+      "loss": 0.6778,
       "step": 14
     },
     {
       "epoch": 0.6,
-      "grad_norm": 0.4762842655181885,
-      "learning_rate": 0.00019774033898178667,
-      "loss": 0.5434,
+      "grad_norm": 0.32996928691864014,
+      "learning_rate": 0.00019458172417006347,
+      "loss": 0.7032,
       "step": 15
     },
     {
       "epoch": 0.64,
-      "grad_norm": 0.4918793737888336,
-      "learning_rate": 0.00019714298932647098,
-      "loss": 0.5295,
+      "grad_norm": 0.32944002747535706,
+      "learning_rate": 0.0001934564464599461,
+      "loss": 0.7088,
       "step": 16
     },
     {
       "epoch": 0.68,
-      "grad_norm": 0.4949023425579071,
-      "learning_rate": 0.0001964767868814516,
-      "loss": 0.5382,
+      "grad_norm": 0.3067278265953064,
+      "learning_rate": 0.00019222897549773848,
+      "loss": 0.6803,
       "step": 17
     },
     {
       "epoch": 0.72,
-      "grad_norm": 0.4618944227695465,
-      "learning_rate": 0.00019574220383620055,
-      "loss": 0.517,
+      "grad_norm": 0.3252066671848297,
+      "learning_rate": 0.00019090065350491626,
+      "loss": 0.6947,
       "step": 18
     },
     {
       "epoch": 0.76,
-      "grad_norm": 0.4404259920120239,
-      "learning_rate": 0.00019493976084683813,
-      "loss": 0.5089,
+      "grad_norm": 0.31279104948043823,
+      "learning_rate": 0.00018947293298207635,
+      "loss": 0.6402,
       "step": 19
     },
     {
       "epoch": 0.8,
-      "grad_norm": 0.47014883160591125,
-      "learning_rate": 0.00019407002666710336,
-      "loss": 0.5412,
+      "grad_norm": 0.30892765522003174,
+      "learning_rate": 0.0001879473751206489,
+      "loss": 0.6519,
       "step": 20
     },
     {
       "epoch": 0.84,
-      "grad_norm": 0.49193090200424194,
-      "learning_rate": 0.00019313361774523385,
-      "loss": 0.5533,
+      "grad_norm": 0.29156580567359924,
+      "learning_rate": 0.00018632564809575742,
+      "loss": 0.6747,
       "step": 21
     },
     {
       "epoch": 0.88,
-      "grad_norm": 0.47287049889564514,
-      "learning_rate": 0.00019213119778704128,
-      "loss": 0.486,
+      "grad_norm": 0.2854658365249634,
+      "learning_rate": 0.00018460952524209355,
+      "loss": 0.6556,
       "step": 22
     },
     {
       "epoch": 0.92,
-      "grad_norm": 0.4395417273044586,
-      "learning_rate": 0.00019106347728549135,
-      "loss": 0.4964,
+      "grad_norm": 0.2657647430896759,
+      "learning_rate": 0.00018280088311480201,
+      "loss": 0.6128,
       "step": 23
     },
     {
       "epoch": 0.96,
-      "grad_norm": 0.4677005708217621,
-      "learning_rate": 0.00018993121301712193,
-      "loss": 0.5101,
+      "grad_norm": 0.29050764441490173,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 0.6674,
       "step": 24
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.4530835449695587,
-      "learning_rate": 0.00018873520750565718,
-      "loss": 0.4903,
+      "grad_norm": 0.29195520281791687,
+      "learning_rate": 0.00017891405093963938,
+      "loss": 0.63,
       "step": 25
     },
     {
       "epoch": 1.04,
-      "grad_norm": 0.41354307532310486,
-      "learning_rate": 0.00018747630845319612,
-      "loss": 0.3837,
+      "grad_norm": 0.30334123969078064,
+      "learning_rate": 0.00017684011108568592,
+      "loss": 0.5533,
       "step": 26
     },
     {
       "epoch": 1.08,
-      "grad_norm": 0.4053184986114502,
-      "learning_rate": 0.0001861554081393806,
-      "loss": 0.3586,
+      "grad_norm": 0.30424660444259644,
+      "learning_rate": 0.0001746821476984154,
+      "loss": 0.5549,
       "step": 27
     },
     {
       "epoch": 1.12,
-      "grad_norm": 0.39625370502471924,
-      "learning_rate": 0.0001847734427889671,
-      "loss": 0.3408,
+      "grad_norm": 0.28369948267936707,
+      "learning_rate": 0.00017244252047910892,
+      "loss": 0.5415,
       "step": 28
     },
     {
       "epoch": 1.16,
-      "grad_norm": 0.3945152759552002,
-      "learning_rate": 0.0001833313919082515,
-      "loss": 0.3329,
+      "grad_norm": 0.2778087258338928,
+      "learning_rate": 0.00017012367842724887,
+      "loss": 0.5382,
       "step": 29
     },
     {
       "epoch": 1.2,
-      "grad_norm": 0.45374399423599243,
-      "learning_rate": 0.0001818302775908169,
-      "loss": 0.3273,
+      "grad_norm": 0.2843495011329651,
+      "learning_rate": 0.00016772815716257412,
+      "loss": 0.5388,
       "step": 30
     },
     {
       "epoch": 1.24,
-      "grad_norm": 0.44107717275619507,
-      "learning_rate": 0.00018027116379309638,
-      "loss": 0.3397,
+      "grad_norm": 0.28178870677948,
+      "learning_rate": 0.00016525857615241687,
+      "loss": 0.5126,
       "step": 31
     },
     {
       "epoch": 1.28,
-      "grad_norm": 0.46182313561439514,
-      "learning_rate": 0.00017865515558026428,
-      "loss": 0.3439,
+      "grad_norm": 0.28643131256103516,
+      "learning_rate": 0.0001627176358473537,
+      "loss": 0.5167,
       "step": 32
     },
     {
       "epoch": 1.32,
-      "grad_norm": 0.4914720356464386,
-      "learning_rate": 0.00017698339834299061,
-      "loss": 0.3415,
+      "grad_norm": 0.28843289613723755,
+      "learning_rate": 0.00016010811472830252,
+      "loss": 0.5099,
       "step": 33
     },
     {
       "epoch": 1.3599999999999999,
-      "grad_norm": 0.46283280849456787,
-      "learning_rate": 0.00017525707698561385,
-      "loss": 0.3309,
+      "grad_norm": 0.28687819838523865,
+      "learning_rate": 0.00015743286626829437,
+      "loss": 0.5185,
       "step": 34
     },
     {
       "epoch": 1.4,
-      "grad_norm": 0.47037574648857117,
-      "learning_rate": 0.00017347741508630672,
-      "loss": 0.3311,
+      "grad_norm": 0.28858163952827454,
+      "learning_rate": 0.00015469481581224272,
+      "loss": 0.5279,
       "step": 35
     },
     {
       "epoch": 1.44,
-      "grad_norm": 0.4362049102783203,
-      "learning_rate": 0.00017164567402983152,
-      "loss": 0.3185,
+      "grad_norm": 0.2918362021446228,
+      "learning_rate": 0.00015189695737812152,
+      "loss": 0.4788,
       "step": 36
     },
     {
       "epoch": 1.48,
-      "grad_norm": 0.4418430030345917,
-      "learning_rate": 0.0001697631521134985,
-      "loss": 0.3189,
+      "grad_norm": 0.291130930185318,
+      "learning_rate": 0.00014904235038305083,
+      "loss": 0.4856,
       "step": 37
     },
     {
       "epoch": 1.52,
-      "grad_norm": 0.4334975779056549,
-      "learning_rate": 0.00016783118362696163,
-      "loss": 0.3152,
+      "grad_norm": 0.2943766415119171,
+      "learning_rate": 0.0001461341162978688,
+      "loss": 0.4615,
       "step": 38
     },
     {
       "epoch": 1.56,
-      "grad_norm": 0.42978766560554504,
-      "learning_rate": 0.00016585113790650388,
-      "loss": 0.3302,
+      "grad_norm": 0.297374427318573,
+      "learning_rate": 0.00014317543523384928,
+      "loss": 0.4583,
       "step": 39
     },
     {
       "epoch": 1.6,
-      "grad_norm": 0.4448607563972473,
-      "learning_rate": 0.00016382441836448202,
-      "loss": 0.3311,
+      "grad_norm": 0.30772677063941956,
+      "learning_rate": 0.00014016954246529696,
+      "loss": 0.4859,
       "step": 40
     },
     {
       "epoch": 1.6400000000000001,
-      "grad_norm": 0.40594780445098877,
-      "learning_rate": 0.0001617524614946192,
-      "loss": 0.326,
+      "grad_norm": 0.29820892214775085,
+      "learning_rate": 0.00013711972489182208,
+      "loss": 0.4768,
       "step": 41
     },
     {
       "epoch": 1.6800000000000002,
-      "grad_norm": 0.4091069996356964,
-      "learning_rate": 0.00015963673585385016,
-      "loss": 0.3374,
+      "grad_norm": 0.29821571707725525,
+      "learning_rate": 0.00013402931744416433,
+      "loss": 0.4891,
       "step": 42
     },
     {
       "epoch": 1.72,
-      "grad_norm": 0.42589232325553894,
-      "learning_rate": 0.0001574787410214407,
-      "loss": 0.3228,
+      "grad_norm": 0.2966136634349823,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 0.4794,
       "step": 43
     },
     {
       "epoch": 1.76,
-      "grad_norm": 0.413335919380188,
-      "learning_rate": 0.00015528000653611935,
-      "loss": 0.3234,
+      "grad_norm": 0.3031865358352661,
+      "learning_rate": 0.00012774029087618446,
+      "loss": 0.4853,
       "step": 44
     },
     {
       "epoch": 1.8,
-      "grad_norm": 0.44703832268714905,
-      "learning_rate": 0.00015304209081197425,
-      "loss": 0.3352,
+      "grad_norm": 0.2918758690357208,
+      "learning_rate": 0.00012454854871407994,
+      "loss": 0.4523,
       "step": 45
     },
     {
       "epoch": 1.8399999999999999,
-      "grad_norm": 0.42445358633995056,
-      "learning_rate": 0.000150766580033884,
-      "loss": 0.3429,
+      "grad_norm": 0.3020419180393219,
+      "learning_rate": 0.0001213299630743747,
+      "loss": 0.4668,
       "step": 46
     },
     {
       "epoch": 1.88,
-      "grad_norm": 0.39593955874443054,
-      "learning_rate": 0.00014845508703326504,
-      "loss": 0.3247,
+      "grad_norm": 0.30048614740371704,
+      "learning_rate": 0.000118088053433211,
+      "loss": 0.4468,
       "step": 47
     },
     {
       "epoch": 1.92,
-      "grad_norm": 0.40969106554985046,
-      "learning_rate": 0.0001461092501449326,
-      "loss": 0.3126,
+      "grad_norm": 0.3022134304046631,
+      "learning_rate": 0.0001148263647711842,
+      "loss": 0.4636,
       "step": 48
     },
     {
       "epoch": 1.96,
-      "grad_norm": 0.4226933717727661,
-      "learning_rate": 0.00014373073204588556,
-      "loss": 0.3184,
+      "grad_norm": 0.3102932572364807,
+      "learning_rate": 0.00011154846369695863,
+      "loss": 0.4762,
       "step": 49
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.4182460308074951,
-      "learning_rate": 0.00014132121857683783,
-      "loss": 0.3263,
+      "grad_norm": 0.29303210973739624,
+      "learning_rate": 0.00010825793454723325,
+      "loss": 0.4109,
       "step": 50
     },
     {
       "epoch": 2.04,
-      "grad_norm": 0.3583109676837921,
-      "learning_rate": 0.00013888241754733208,
-      "loss": 0.2484,
+      "grad_norm": 0.29739847779273987,
+      "learning_rate": 0.00010495837546732224,
+      "loss": 0.3787,
       "step": 51
     },
     {
       "epoch": 2.08,
-      "grad_norm": 0.3850906491279602,
-      "learning_rate": 0.00013641605752528224,
-      "loss": 0.24,
+      "grad_norm": 0.29793456196784973,
+      "learning_rate": 0.00010165339447663587,
+      "loss": 0.3708,
       "step": 52
     },
     {
       "epoch": 2.12,
-      "grad_norm": 0.3770233392715454,
-      "learning_rate": 0.00013392388661180303,
-      "loss": 0.2334,
+      "grad_norm": 0.31656354665756226,
+      "learning_rate": 9.834660552336415e-05,
+      "loss": 0.3855,
       "step": 53
     },
     {
       "epoch": 2.16,
-      "grad_norm": 0.374737024307251,
-      "learning_rate": 0.0001314076712021949,
-      "loss": 0.244,
+      "grad_norm": 0.30371928215026855,
+      "learning_rate": 9.504162453267777e-05,
+      "loss": 0.3873,
       "step": 54
     },
     {
       "epoch": 2.2,
-      "grad_norm": 0.37868672609329224,
-      "learning_rate": 0.0001288691947339621,
-      "loss": 0.2134,
+      "grad_norm": 0.31064578890800476,
+      "learning_rate": 9.174206545276677e-05,
+      "loss": 0.3595,
       "step": 55
     },
     {
       "epoch": 2.24,
-      "grad_norm": 0.41098785400390625,
-      "learning_rate": 0.00012631025642275212,
-      "loss": 0.2373,
+      "grad_norm": 0.29833751916885376,
+      "learning_rate": 8.845153630304139e-05,
+      "loss": 0.366,
       "step": 56
     },
     {
       "epoch": 2.2800000000000002,
-      "grad_norm": 0.3870932161808014,
-      "learning_rate": 0.0001237326699871115,
-      "loss": 0.2238,
+      "grad_norm": 0.3072323501110077,
+      "learning_rate": 8.517363522881579e-05,
+      "loss": 0.3621,
       "step": 57
     },
     {
       "epoch": 2.32,
-      "grad_norm": 0.4083153307437897,
-      "learning_rate": 0.00012113826236296244,
-      "loss": 0.2141,
+      "grad_norm": 0.29597151279449463,
+      "learning_rate": 8.191194656678904e-05,
+      "loss": 0.3519,
       "step": 58
     },
     {
       "epoch": 2.36,
-      "grad_norm": 0.412580668926239,
-      "learning_rate": 0.00011852887240871145,
-      "loss": 0.2201,
+      "grad_norm": 0.3137385845184326,
+      "learning_rate": 7.867003692562534e-05,
+      "loss": 0.3519,
       "step": 59
     },
     {
       "epoch": 2.4,
-      "grad_norm": 0.4499835968017578,
-      "learning_rate": 0.00011590634960190721,
-      "loss": 0.2215,
+      "grad_norm": 0.3161297142505646,
+      "learning_rate": 7.54514512859201e-05,
+      "loss": 0.3502,
       "step": 60
     },
     {
       "epoch": 2.44,
-      "grad_norm": 0.48053982853889465,
-      "learning_rate": 0.00011327255272837221,
-      "loss": 0.217,
+      "grad_norm": 0.3286161720752716,
+      "learning_rate": 7.225970912381556e-05,
+      "loss": 0.3518,
       "step": 61
     },
     {
       "epoch": 2.48,
-      "grad_norm": 0.44269296526908875,
-      "learning_rate": 0.00011062934856473655,
-      "loss": 0.2212,
+      "grad_norm": 0.3324306905269623,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 0.357,
       "step": 62
     },
     {
       "epoch": 2.52,
-      "grad_norm": 0.4232514202594757,
-      "learning_rate": 0.00010797861055530831,
-      "loss": 0.2289,
+      "grad_norm": 0.3510464131832123,
+      "learning_rate": 6.59706825558357e-05,
+      "loss": 0.379,
       "step": 63
     },
     {
       "epoch": 2.56,
-      "grad_norm": 0.41652482748031616,
-      "learning_rate": 0.00010532221748421787,
-      "loss": 0.2098,
+      "grad_norm": 0.3358787000179291,
+      "learning_rate": 6.28802751081779e-05,
+      "loss": 0.3399,
       "step": 64
     },
     {
       "epoch": 2.6,
-      "grad_norm": 0.39650851488113403,
-      "learning_rate": 0.00010266205214377748,
-      "loss": 0.195,
+      "grad_norm": 0.330174058675766,
+      "learning_rate": 5.983045753470308e-05,
+      "loss": 0.3465,
       "step": 65
     },
     {
       "epoch": 2.64,
-      "grad_norm": 0.39610394835472107,
-      "learning_rate": 0.0001,
-      "loss": 0.1996,
+      "grad_norm": 0.31485724449157715,
+      "learning_rate": 5.6824564766150726e-05,
+      "loss": 0.3249,
       "step": 66
     },
     {
       "epoch": 2.68,
-      "grad_norm": 0.3933280110359192,
-      "learning_rate": 9.733794785622253e-05,
-      "loss": 0.2243,
+      "grad_norm": 0.3314495384693146,
+      "learning_rate": 5.386588370213124e-05,
+      "loss": 0.3565,
       "step": 67
     },
     {
       "epoch": 2.7199999999999998,
-      "grad_norm": 0.3946898579597473,
-      "learning_rate": 9.467778251578217e-05,
-      "loss": 0.2216,
+      "grad_norm": 0.334344744682312,
+      "learning_rate": 5.095764961694922e-05,
+      "loss": 0.344,
       "step": 68
     },
     {
       "epoch": 2.76,
-      "grad_norm": 0.4111415147781372,
-      "learning_rate": 9.202138944469168e-05,
-      "loss": 0.2261,
+      "grad_norm": 0.330770343542099,
+      "learning_rate": 4.810304262187852e-05,
+      "loss": 0.3424,
       "step": 69
     },
     {
       "epoch": 2.8,
-      "grad_norm": 0.4122645854949951,
-      "learning_rate": 8.937065143526347e-05,
-      "loss": 0.226,
+      "grad_norm": 0.32071828842163086,
+      "learning_rate": 4.530518418775733e-05,
+      "loss": 0.3396,
       "step": 70
     },
     {
       "epoch": 2.84,
-      "grad_norm": 0.4040912091732025,
-      "learning_rate": 8.672744727162781e-05,
-      "loss": 0.2397,
+      "grad_norm": 0.32281526923179626,
+      "learning_rate": 4.256713373170564e-05,
+      "loss": 0.343,
       "step": 71
     },
     {
       "epoch": 2.88,
-      "grad_norm": 0.4039236605167389,
-      "learning_rate": 8.409365039809281e-05,
-      "loss": 0.214,
+      "grad_norm": 0.3217308223247528,
+      "learning_rate": 3.9891885271697496e-05,
+      "loss": 0.3142,
       "step": 72
     },
     {
       "epoch": 2.92,
-      "grad_norm": 0.39346644282341003,
-      "learning_rate": 8.147112759128859e-05,
-      "loss": 0.2211,
+      "grad_norm": 0.3167681097984314,
+      "learning_rate": 3.7282364152646297e-05,
+      "loss": 0.3351,
       "step": 73
     },
     {
       "epoch": 2.96,
-      "grad_norm": 0.4206030070781708,
-      "learning_rate": 7.886173763703757e-05,
-      "loss": 0.2193,
+      "grad_norm": 0.3280259072780609,
+      "learning_rate": 3.4741423847583134e-05,
+      "loss": 0.316,
       "step": 74
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.3998810648918152,
-      "learning_rate": 7.626733001288851e-05,
-      "loss": 0.2129,
+      "grad_norm": 0.32448863983154297,
+      "learning_rate": 3.227184283742591e-05,
+      "loss": 0.3195,
       "step": 75
     },
     {
       "epoch": 3.04,
-      "grad_norm": 0.3447999060153961,
-      "learning_rate": 7.368974357724789e-05,
-      "loss": 0.1809,
+      "grad_norm": 0.33512407541275024,
+      "learning_rate": 2.9876321572751144e-05,
+      "loss": 0.2814,
       "step": 76
     },
     {
       "epoch": 3.08,
-      "grad_norm": 0.3176288306713104,
-      "learning_rate": 7.113080526603792e-05,
-      "loss": 0.1648,
+      "grad_norm": 0.33603280782699585,
+      "learning_rate": 2.7557479520891104e-05,
+      "loss": 0.2981,
       "step": 77
     },
     {
       "epoch": 3.12,
-      "grad_norm": 0.32004985213279724,
-      "learning_rate": 6.859232879780515e-05,
-      "loss": 0.1582,
+      "grad_norm": 0.33797961473464966,
+      "learning_rate": 2.5317852301584643e-05,
+      "loss": 0.293,
       "step": 78
     },
     {
       "epoch": 3.16,
-      "grad_norm": 0.31019288301467896,
-      "learning_rate": 6.607611338819697e-05,
-      "loss": 0.1502,
+      "grad_norm": 0.3338729739189148,
+      "learning_rate": 2.315988891431412e-05,
+      "loss": 0.2988,
       "step": 79
     },
     {
       "epoch": 3.2,
-      "grad_norm": 0.31684502959251404,
-      "learning_rate": 6.358394247471778e-05,
-      "loss": 0.1514,
+      "grad_norm": 0.3124498128890991,
+      "learning_rate": 2.1085949060360654e-05,
+      "loss": 0.2956,
       "step": 80
     },
     {
       "epoch": 3.24,
-      "grad_norm": 0.3580498695373535,
-      "learning_rate": 6.111758245266794e-05,
-      "loss": 0.1533,
+      "grad_norm": 0.30723869800567627,
+      "learning_rate": 1.9098300562505266e-05,
+      "loss": 0.2762,
       "step": 81
     },
     {
       "epoch": 3.2800000000000002,
-      "grad_norm": 0.3377327024936676,
-      "learning_rate": 5.867878142316221e-05,
-      "loss": 0.1522,
+      "grad_norm": 0.30877774953842163,
+      "learning_rate": 1.7199116885197995e-05,
+      "loss": 0.2811,
       "step": 82
     },
     {
       "epoch": 3.32,
-      "grad_norm": 0.36469805240631104,
-      "learning_rate": 5.626926795411447e-05,
-      "loss": 0.1625,
+      "grad_norm": 0.3040075898170471,
+      "learning_rate": 1.5390474757906446e-05,
+      "loss": 0.2843,
       "step": 83
     },
     {
       "epoch": 3.36,
-      "grad_norm": 0.3576640486717224,
-      "learning_rate": 5.38907498550674e-05,
-      "loss": 0.1491,
+      "grad_norm": 0.29099854826927185,
+      "learning_rate": 1.3674351904242611e-05,
+      "loss": 0.2823,
       "step": 84
     },
     {
       "epoch": 3.4,
-      "grad_norm": 0.36043015122413635,
-      "learning_rate": 5.1544912966734994e-05,
-      "loss": 0.1561,
+      "grad_norm": 0.29945293068885803,
+      "learning_rate": 1.2052624879351104e-05,
+      "loss": 0.3007,
       "step": 85
     },
     {
       "epoch": 3.44,
-      "grad_norm": 0.3925049602985382,
-      "learning_rate": 4.9233419966116036e-05,
-      "loss": 0.1526,
+      "grad_norm": 0.29691964387893677,
+      "learning_rate": 1.0527067017923654e-05,
+      "loss": 0.2866,
       "step": 86
     },
     {
       "epoch": 3.48,
-      "grad_norm": 0.3896467685699463,
-      "learning_rate": 4.695790918802576e-05,
-      "loss": 0.1626,
+      "grad_norm": 0.29864245653152466,
+      "learning_rate": 9.09934649508375e-06,
+      "loss": 0.2946,
       "step": 87
     },
     {
       "epoch": 3.52,
-      "grad_norm": 0.38003793358802795,
-      "learning_rate": 4.47199934638807e-05,
-      "loss": 0.1483,
+      "grad_norm": 0.29509297013282776,
+      "learning_rate": 7.771024502261526e-06,
+      "loss": 0.2717,
       "step": 88
     },
     {
       "epoch": 3.56,
-      "grad_norm": 0.3852701187133789,
-      "learning_rate": 4.252125897855932e-05,
-      "loss": 0.1494,
+      "grad_norm": 0.3103945851325989,
+      "learning_rate": 6.543553540053926e-06,
+      "loss": 0.287,
       "step": 89
     },
     {
       "epoch": 3.6,
-      "grad_norm": 0.38289740681648254,
-      "learning_rate": 4.036326414614985e-05,
-      "loss": 0.1694,
+      "grad_norm": 0.2974102795124054,
+      "learning_rate": 5.418275829936537e-06,
+      "loss": 0.2815,
       "step": 90
     },
     {
       "epoch": 3.64,
-      "grad_norm": 0.3584575057029724,
-      "learning_rate": 3.824753850538082e-05,
-      "loss": 0.1581,
+      "grad_norm": 0.3041694760322571,
+      "learning_rate": 4.3964218465642355e-06,
+      "loss": 0.2905,
       "step": 91
     },
     {
       "epoch": 3.68,
-      "grad_norm": 0.36479833722114563,
-      "learning_rate": 3.617558163551802e-05,
-      "loss": 0.1567,
+      "grad_norm": 0.2940130829811096,
+      "learning_rate": 3.4791089722651436e-06,
+      "loss": 0.273,
       "step": 92
     },
     {
       "epoch": 3.7199999999999998,
-      "grad_norm": 0.3648757338523865,
-      "learning_rate": 3.414886209349615e-05,
-      "loss": 0.1539,
+      "grad_norm": 0.2969824969768524,
+      "learning_rate": 2.667340275199426e-06,
+      "loss": 0.2691,
       "step": 93
     },
     {
       "epoch": 3.76,
-      "grad_norm": 0.3735385835170746,
-      "learning_rate": 3.216881637303839e-05,
-      "loss": 0.1455,
+      "grad_norm": 0.30998706817626953,
+      "learning_rate": 1.9620034125190644e-06,
+      "loss": 0.2984,
       "step": 94
     },
     {
       "epoch": 3.8,
-      "grad_norm": 0.3788692355155945,
-      "learning_rate": 3.0236847886501542e-05,
-      "loss": 0.1512,
+      "grad_norm": 0.31238672137260437,
+      "learning_rate": 1.3638696597277679e-06,
+      "loss": 0.2856,
       "step": 95
     },
     {
       "epoch": 3.84,
-      "grad_norm": 0.36534735560417175,
-      "learning_rate": 2.8354325970168484e-05,
-      "loss": 0.1539,
+      "grad_norm": 0.29781845211982727,
+      "learning_rate": 8.735930673024806e-07,
+      "loss": 0.2911,
       "step": 96
     },
     {
       "epoch": 3.88,
-      "grad_norm": 0.3561382591724396,
-      "learning_rate": 2.6522584913693294e-05,
-      "loss": 0.1613,
+      "grad_norm": 0.29917433857917786,
+      "learning_rate": 4.917097454988584e-07,
+      "loss": 0.278,
       "step": 97
     },
     {
       "epoch": 3.92,
-      "grad_norm": 0.36043405532836914,
-      "learning_rate": 2.4742923014386156e-05,
-      "loss": 0.1399,
+      "grad_norm": 0.30273205041885376,
+      "learning_rate": 2.1863727812254653e-07,
+      "loss": 0.2811,
       "step": 98
     },
     {
       "epoch": 3.96,
-      "grad_norm": 0.3461771011352539,
-      "learning_rate": 2.301660165700936e-05,
-      "loss": 0.1446,
+      "grad_norm": 0.3171875774860382,
+      "learning_rate": 5.467426590739511e-08,
+      "loss": 0.2876,
       "step": 99
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.3610660135746002,
-      "learning_rate": 2.1344844419735755e-05,
-      "loss": 0.1669,
-      "step": 100
-    },
-    {
-      "epoch": 4.04,
-      "grad_norm": 0.3192867636680603,
-      "learning_rate": 1.9728836206903656e-05,
-      "loss": 0.138,
-      "step": 101
-    },
-    {
-      "epoch": 4.08,
-      "grad_norm": 0.2961142957210541,
-      "learning_rate": 1.8169722409183097e-05,
-      "loss": 0.1251,
-      "step": 102
-    },
-    {
-      "epoch": 4.12,
-      "grad_norm": 0.33224716782569885,
-      "learning_rate": 1.6668608091748495e-05,
-      "loss": 0.1307,
-      "step": 103
-    },
-    {
-      "epoch": 4.16,
-      "grad_norm": 0.30373403429985046,
-      "learning_rate": 1.522655721103291e-05,
-      "loss": 0.1259,
-      "step": 104
-    },
-    {
-      "epoch": 4.2,
-      "grad_norm": 0.31472688913345337,
-      "learning_rate": 1.3844591860619383e-05,
-      "loss": 0.1267,
-      "step": 105
-    },
-    {
-      "epoch": 4.24,
-      "grad_norm": 0.3078845739364624,
-      "learning_rate": 1.2523691546803873e-05,
-      "loss": 0.123,
-      "step": 106
-    },
-    {
-      "epoch": 4.28,
-      "grad_norm": 0.31925857067108154,
-      "learning_rate": 1.1264792494342857e-05,
-      "loss": 0.1326,
-      "step": 107
-    },
-    {
-      "epoch": 4.32,
-      "grad_norm": 0.29572564363479614,
-      "learning_rate": 1.0068786982878087e-05,
-      "loss": 0.1162,
-      "step": 108
-    },
-    {
-      "epoch": 4.36,
-      "grad_norm": 0.29332664608955383,
-      "learning_rate": 8.936522714508678e-06,
-      "loss": 0.1153,
-      "step": 109
-    },
-    {
-      "epoch": 4.4,
-      "grad_norm": 0.29813987016677856,
-      "learning_rate": 7.868802212958703e-06,
-      "loss": 0.1166,
-      "step": 110
-    },
-    {
-      "epoch": 4.44,
-      "grad_norm": 0.2909725308418274,
-      "learning_rate": 6.866382254766157e-06,
-      "loss": 0.108,
-      "step": 111
-    },
-    {
-      "epoch": 4.48,
-      "grad_norm": 0.31103241443634033,
-      "learning_rate": 5.929973332896677e-06,
-      "loss": 0.1228,
-      "step": 112
-    },
-    {
-      "epoch": 4.52,
-      "grad_norm": 0.29073992371559143,
-      "learning_rate": 5.060239153161872e-06,
-      "loss": 0.118,
-      "step": 113
-    },
-    {
-      "epoch": 4.5600000000000005,
-      "grad_norm": 0.288641095161438,
-      "learning_rate": 4.257796163799455e-06,
-      "loss": 0.1184,
-      "step": 114
-    },
-    {
-      "epoch": 4.6,
-      "grad_norm": 0.3088034689426422,
-      "learning_rate": 3.5232131185484076e-06,
-      "loss": 0.1285,
-      "step": 115
-    },
-    {
-      "epoch": 4.64,
-      "grad_norm": 0.30730554461479187,
-      "learning_rate": 2.857010673529015e-06,
-      "loss": 0.1263,
-      "step": 116
-    },
-    {
-      "epoch": 4.68,
-      "grad_norm": 0.291875422000885,
-      "learning_rate": 2.259661018213333e-06,
-      "loss": 0.108,
-      "step": 117
-    },
-    {
-      "epoch": 4.72,
-      "grad_norm": 0.31049686670303345,
-      "learning_rate": 1.7315875407479032e-06,
-      "loss": 0.1225,
-      "step": 118
-    },
-    {
-      "epoch": 4.76,
-      "grad_norm": 0.32437863945961,
-      "learning_rate": 1.2731645278655445e-06,
-      "loss": 0.1431,
-      "step": 119
-    },
-    {
-      "epoch": 4.8,
-      "grad_norm": 0.3190011978149414,
-      "learning_rate": 8.847168995992916e-07,
-      "loss": 0.1298,
-      "step": 120
-    },
-    {
-      "epoch": 4.84,
-      "grad_norm": 0.3092647194862366,
-      "learning_rate": 5.665199789862907e-07,
-      "loss": 0.1333,
-      "step": 121
-    },
-    {
-      "epoch": 4.88,
-      "grad_norm": 0.3259420096874237,
-      "learning_rate": 3.1879929692498757e-07,
-      "loss": 0.1312,
-      "step": 122
-    },
-    {
-      "epoch": 4.92,
-      "grad_norm": 0.31257393956184387,
-      "learning_rate": 1.4173043232380557e-07,
-      "loss": 0.1255,
-      "step": 123
-    },
-    {
-      "epoch": 4.96,
-      "grad_norm": 0.310062438249588,
-      "learning_rate": 3.5438887654737355e-08,
-      "loss": 0.1155,
-      "step": 124
-    },
-    {
-      "epoch": 5.0,
-      "grad_norm": 0.3081853687763214,
+      "grad_norm": 0.3021162450313568,
       "learning_rate": 0.0,
-      "loss": 0.1245,
-      "step": 125
+      "loss": 0.2926,
+      "step": 100
     }
   ],
   "logging_steps": 1,
-  "max_steps": 125,
+  "max_steps": 100,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -901,7 +726,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.8813347440214016e+17,
+  "total_flos": 1.5000238750511923e+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 sha256:77e1d32de1ebecb4b037f065430ceae70509727e789a1fec0ab879aa962ea3fb
+oid sha256:07de7a446103efc6abf630ef72bd582d5c357a7607d0c9fdc45dd38d04c8d45d
 size 5240