Upload folder using huggingface_hub

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:916af11d0703123ba667e1f3fb7f7f309d584a8ed4448932d2908e2c7e88013c
+oid sha256:f5b13460b33d2a7f490c08d9a5f968391789cb17181c0f5dd613b0c3a5ab4278
 size 134235048

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f6b06f28f0388c0b61c7bbd24012e36baa3d39a961dee7d0dc03afab6d4b891
+oid sha256:f6be95aa7e9079dba8b5c1ff292358256dbe2e50c1edb03e84223a2bf873667d
 size 268543610

special_tokens_map.json

@@ -14,7 +14,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "[PAD]",
+    "content": "</s>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer_config.json

@@ -26,14 +26,6 @@
       "rstrip": false,
       "single_word": false,
       "special": true
-    },
-    "32000": {
-      "content": "[PAD]",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "bos_token": "<s>",
@@ -42,7 +34,7 @@
   "extra_special_tokens": {},
   "legacy": false,
   "model_max_length": 512,
-  "pad_token": "[PAD]",
+  "pad_token": "</s>",
   "padding_side": "right",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,

trainer_state.json

@@ -10,1696 +10,1696 @@
   "log_history": [
     {
       "epoch": 0.012364760432766615,
-      "grad_norm": 0.04875350371003151,
+      "grad_norm": 0.04910803958773613,
       "learning_rate": 0.00029876237623762373,
-      "loss": 1.0085,
+      "loss": 1.0084,
       "step": 10
     },
     {
       "epoch": 0.02472952086553323,
-      "grad_norm": 0.03758955001831055,
+      "grad_norm": 0.037698596715927124,
       "learning_rate": 0.0002975247524752475,
-      "loss": 0.9611,
+      "loss": 0.9609,
       "step": 20
     },
     {
       "epoch": 0.03709428129829984,
-      "grad_norm": 0.036796920001506805,
+      "grad_norm": 0.03782161325216293,
       "learning_rate": 0.00029628712871287126,
       "loss": 0.9033,
       "step": 30
     },
     {
       "epoch": 0.04945904173106646,
-      "grad_norm": 0.04110224172472954,
+      "grad_norm": 0.04191258177161217,
       "learning_rate": 0.000295049504950495,
       "loss": 0.9154,
       "step": 40
     },
     {
       "epoch": 0.061823802163833076,
-      "grad_norm": 0.03811247646808624,
+      "grad_norm": 0.0391441248357296,
       "learning_rate": 0.0002938118811881188,
       "loss": 0.9001,
       "step": 50
     },
     {
       "epoch": 0.07418856259659969,
-      "grad_norm": 0.0396280363202095,
+      "grad_norm": 0.04036989435553551,
       "learning_rate": 0.00029257425742574254,
       "loss": 0.9069,
       "step": 60
     },
     {
       "epoch": 0.0865533230293663,
-      "grad_norm": 0.036311160773038864,
+      "grad_norm": 0.036793895065784454,
       "learning_rate": 0.0002913366336633663,
-      "loss": 0.8905,
+      "loss": 0.8903,
       "step": 70
     },
     {
       "epoch": 0.09891808346213292,
-      "grad_norm": 0.04230085760354996,
+      "grad_norm": 0.04941694810986519,
       "learning_rate": 0.00029009900990099006,
       "loss": 0.928,
       "step": 80
     },
     {
       "epoch": 0.11128284389489954,
-      "grad_norm": 0.03857972100377083,
+      "grad_norm": 0.03952586278319359,
       "learning_rate": 0.0002888613861386138,
-      "loss": 0.9122,
+      "loss": 0.9121,
       "step": 90
     },
     {
       "epoch": 0.12364760432766615,
-      "grad_norm": 0.043333932757377625,
+      "grad_norm": 0.04417801648378372,
       "learning_rate": 0.0002876237623762376,
-      "loss": 0.8921,
+      "loss": 0.8918,
       "step": 100
     },
     {
       "epoch": 0.13601236476043277,
-      "grad_norm": 0.03789997100830078,
+      "grad_norm": 0.03843015059828758,
       "learning_rate": 0.00028638613861386135,
       "loss": 0.8988,
       "step": 110
     },
     {
       "epoch": 0.14837712519319937,
-      "grad_norm": 0.03549322113394737,
+      "grad_norm": 0.035807665437459946,
       "learning_rate": 0.0002851485148514851,
       "loss": 0.899,
       "step": 120
     },
     {
       "epoch": 0.160741885625966,
-      "grad_norm": 0.04001789167523384,
+      "grad_norm": 0.041015319526195526,
       "learning_rate": 0.00028391089108910887,
       "loss": 0.8937,
       "step": 130
     },
     {
       "epoch": 0.1731066460587326,
-      "grad_norm": 0.04051420837640762,
+      "grad_norm": 0.04108859598636627,
       "learning_rate": 0.00028267326732673263,
-      "loss": 0.8941,
+      "loss": 0.894,
       "step": 140
     },
     {
       "epoch": 0.18547140649149924,
-      "grad_norm": 0.03879082202911377,
+      "grad_norm": 0.03874868154525757,
       "learning_rate": 0.0002814356435643564,
-      "loss": 0.9086,
+      "loss": 0.9088,
       "step": 150
     },
     {
       "epoch": 0.19783616692426584,
-      "grad_norm": 0.03938911855220795,
+      "grad_norm": 0.03931848704814911,
       "learning_rate": 0.00028019801980198015,
-      "loss": 0.908,
+      "loss": 0.9079,
       "step": 160
     },
     {
       "epoch": 0.21020092735703247,
-      "grad_norm": 0.04576217010617256,
+      "grad_norm": 0.04638601467013359,
       "learning_rate": 0.0002789603960396039,
-      "loss": 0.9058,
+      "loss": 0.9057,
       "step": 170
     },
     {
       "epoch": 0.22256568778979907,
-      "grad_norm": 0.05652037635445595,
+      "grad_norm": 0.05649528279900551,
       "learning_rate": 0.00027772277227722773,
-      "loss": 0.8718,
+      "loss": 0.872,
       "step": 180
     },
     {
       "epoch": 0.23493044822256567,
-      "grad_norm": 0.04269680753350258,
+      "grad_norm": 0.043013814836740494,
       "learning_rate": 0.00027648514851485144,
       "loss": 0.9102,
       "step": 190
     },
     {
       "epoch": 0.2472952086553323,
-      "grad_norm": 0.046770963817834854,
+      "grad_norm": 0.04763510450720787,
       "learning_rate": 0.0002752475247524752,
-      "loss": 0.8893,
+      "loss": 0.8894,
       "step": 200
     },
     {
       "epoch": 0.2596599690880989,
-      "grad_norm": 0.0391731895506382,
+      "grad_norm": 0.03904448449611664,
       "learning_rate": 0.000274009900990099,
-      "loss": 0.8858,
+      "loss": 0.8856,
       "step": 210
     },
     {
       "epoch": 0.27202472952086554,
-      "grad_norm": 0.04302387312054634,
+      "grad_norm": 0.04328664019703865,
       "learning_rate": 0.0002727722772277227,
       "loss": 0.8967,
       "step": 220
     },
     {
       "epoch": 0.28438948995363217,
-      "grad_norm": 0.04622489586472511,
+      "grad_norm": 0.04718885198235512,
       "learning_rate": 0.0002715346534653465,
-      "loss": 0.8919,
+      "loss": 0.892,
       "step": 230
     },
     {
       "epoch": 0.29675425038639874,
-      "grad_norm": 0.04950324073433876,
+      "grad_norm": 0.048305947333574295,
       "learning_rate": 0.0002702970297029703,
-      "loss": 0.8898,
+      "loss": 0.8896,
       "step": 240
     },
     {
       "epoch": 0.3091190108191654,
-      "grad_norm": 0.047146428376436234,
+      "grad_norm": 0.04656008258461952,
       "learning_rate": 0.000269059405940594,
-      "loss": 0.8941,
+      "loss": 0.8942,
       "step": 250
     },
     {
       "epoch": 0.321483771251932,
-      "grad_norm": 0.04186677187681198,
+      "grad_norm": 0.041436970233917236,
       "learning_rate": 0.00026782178217821777,
       "loss": 0.892,
       "step": 260
     },
     {
       "epoch": 0.33384853168469864,
-      "grad_norm": 0.04358995333313942,
+      "grad_norm": 0.043510112911462784,
       "learning_rate": 0.0002665841584158416,
       "loss": 0.8813,
       "step": 270
     },
     {
       "epoch": 0.3462132921174652,
-      "grad_norm": 0.03836526349186897,
+      "grad_norm": 0.039642177522182465,
       "learning_rate": 0.00026534653465346534,
-      "loss": 0.8883,
+      "loss": 0.8884,
       "step": 280
     },
     {
       "epoch": 0.35857805255023184,
-      "grad_norm": 0.04279692843556404,
+      "grad_norm": 0.043365489691495895,
       "learning_rate": 0.00026410891089108905,
-      "loss": 0.8964,
+      "loss": 0.8963,
       "step": 290
     },
     {
       "epoch": 0.37094281298299847,
-      "grad_norm": 0.04259683936834335,
+      "grad_norm": 0.04440496489405632,
       "learning_rate": 0.00026287128712871287,
-      "loss": 0.8864,
+      "loss": 0.8865,
       "step": 300
     },
     {
       "epoch": 0.38330757341576505,
-      "grad_norm": 0.04212690517306328,
+      "grad_norm": 0.04222091659903526,
       "learning_rate": 0.00026163366336633663,
-      "loss": 0.8937,
+      "loss": 0.8935,
       "step": 310
     },
     {
       "epoch": 0.3956723338485317,
-      "grad_norm": 0.04193605110049248,
+      "grad_norm": 0.04118286073207855,
       "learning_rate": 0.00026039603960396033,
-      "loss": 0.8743,
+      "loss": 0.8742,
       "step": 320
     },
     {
       "epoch": 0.4080370942812983,
-      "grad_norm": 0.043696388602256775,
+      "grad_norm": 0.04324512556195259,
       "learning_rate": 0.00025915841584158415,
-      "loss": 0.8919,
+      "loss": 0.8918,
       "step": 330
     },
     {
       "epoch": 0.42040185471406494,
-      "grad_norm": 0.04463732987642288,
+      "grad_norm": 0.04111215099692345,
       "learning_rate": 0.0002579207920792079,
-      "loss": 0.8791,
+      "loss": 0.8795,
       "step": 340
     },
     {
       "epoch": 0.4327666151468315,
-      "grad_norm": 0.042219433933496475,
+      "grad_norm": 0.04449688270688057,
       "learning_rate": 0.0002566831683168316,
-      "loss": 0.8886,
+      "loss": 0.8889,
       "step": 350
     },
     {
       "epoch": 0.44513137557959814,
-      "grad_norm": 0.04634915664792061,
+      "grad_norm": 0.04658028110861778,
       "learning_rate": 0.00025544554455445543,
       "loss": 0.8798,
       "step": 360
     },
     {
       "epoch": 0.4574961360123648,
-      "grad_norm": 0.03766421601176262,
+      "grad_norm": 0.037669096142053604,
       "learning_rate": 0.0002542079207920792,
       "loss": 0.8809,
       "step": 370
     },
     {
       "epoch": 0.46986089644513135,
-      "grad_norm": 0.04153716191649437,
+      "grad_norm": 0.04158737137913704,
       "learning_rate": 0.00025297029702970296,
-      "loss": 0.8921,
+      "loss": 0.8922,
       "step": 380
     },
     {
       "epoch": 0.482225656877898,
-      "grad_norm": 0.04694748297333717,
+      "grad_norm": 0.047567520290613174,
       "learning_rate": 0.0002517326732673267,
-      "loss": 0.895,
+      "loss": 0.8949,
       "step": 390
     },
     {
       "epoch": 0.4945904173106646,
-      "grad_norm": 0.05713290721178055,
+      "grad_norm": 0.057194869965314865,
       "learning_rate": 0.0002504950495049505,
       "loss": 0.8774,
       "step": 400
     },
     {
       "epoch": 0.5069551777434312,
-      "grad_norm": 0.0414641909301281,
+      "grad_norm": 0.04181217402219772,
       "learning_rate": 0.00024925742574257424,
-      "loss": 0.8908,
+      "loss": 0.8907,
       "step": 410
     },
     {
       "epoch": 0.5193199381761978,
-      "grad_norm": 0.04552585631608963,
+      "grad_norm": 0.045876242220401764,
       "learning_rate": 0.000248019801980198,
-      "loss": 0.8843,
+      "loss": 0.8841,
       "step": 420
     },
     {
       "epoch": 0.5316846986089645,
-      "grad_norm": 0.04167173057794571,
+      "grad_norm": 0.041932158172130585,
       "learning_rate": 0.00024678217821782176,
-      "loss": 0.8583,
+      "loss": 0.8582,
       "step": 430
     },
     {
       "epoch": 0.5440494590417311,
-      "grad_norm": 0.04508620873093605,
+      "grad_norm": 0.044740475714206696,
       "learning_rate": 0.0002455445544554455,
-      "loss": 0.9205,
+      "loss": 0.9204,
       "step": 440
     },
     {
       "epoch": 0.5564142194744977,
-      "grad_norm": 0.04546656087040901,
+      "grad_norm": 0.04608389362692833,
       "learning_rate": 0.0002443069306930693,
-      "loss": 0.8857,
+      "loss": 0.8859,
       "step": 450
     },
     {
       "epoch": 0.5687789799072643,
-      "grad_norm": 0.046972740441560745,
+      "grad_norm": 0.04750910773873329,
       "learning_rate": 0.00024306930693069305,
-      "loss": 0.8788,
+      "loss": 0.8786,
       "step": 460
     },
     {
       "epoch": 0.5811437403400309,
-      "grad_norm": 0.03991515189409256,
+      "grad_norm": 0.0407898910343647,
       "learning_rate": 0.0002418316831683168,
       "loss": 0.8731,
       "step": 470
     },
     {
       "epoch": 0.5935085007727975,
-      "grad_norm": 0.047520458698272705,
+      "grad_norm": 0.047790151089429855,
       "learning_rate": 0.0002405940594059406,
-      "loss": 0.8986,
+      "loss": 0.8984,
       "step": 480
     },
     {
       "epoch": 0.6058732612055642,
-      "grad_norm": 0.04166582226753235,
+      "grad_norm": 0.04182444140315056,
       "learning_rate": 0.00023935643564356433,
       "loss": 0.8865,
       "step": 490
     },
     {
       "epoch": 0.6182380216383307,
-      "grad_norm": 0.043145373463630676,
+      "grad_norm": 0.04374885559082031,
       "learning_rate": 0.0002381188118811881,
-      "loss": 0.8738,
+      "loss": 0.8736,
       "step": 500
     },
     {
       "epoch": 0.6306027820710973,
-      "grad_norm": 0.04514694958925247,
+      "grad_norm": 0.04540247470140457,
       "learning_rate": 0.00023688118811881188,
-      "loss": 0.8976,
+      "loss": 0.8977,
       "step": 510
     },
     {
       "epoch": 0.642967542503864,
-      "grad_norm": 0.03927430510520935,
+      "grad_norm": 0.039125751703977585,
       "learning_rate": 0.00023564356435643561,
-      "loss": 0.8953,
+      "loss": 0.8955,
       "step": 520
     },
     {
       "epoch": 0.6553323029366306,
-      "grad_norm": 0.048136577010154724,
+      "grad_norm": 0.04842868447303772,
       "learning_rate": 0.00023440594059405938,
       "loss": 0.8979,
       "step": 530
     },
     {
       "epoch": 0.6676970633693973,
-      "grad_norm": 0.043889183551073074,
+      "grad_norm": 0.04414287582039833,
       "learning_rate": 0.00023316831683168316,
-      "loss": 0.9065,
+      "loss": 0.9063,
       "step": 540
     },
     {
       "epoch": 0.6800618238021638,
-      "grad_norm": 0.05049331858754158,
+      "grad_norm": 0.05018250271677971,
       "learning_rate": 0.0002319306930693069,
-      "loss": 0.8941,
+      "loss": 0.894,
       "step": 550
     },
     {
       "epoch": 0.6924265842349304,
-      "grad_norm": 0.04710015654563904,
+      "grad_norm": 0.04726792126893997,
       "learning_rate": 0.00023069306930693066,
       "loss": 0.8716,
       "step": 560
     },
     {
       "epoch": 0.7047913446676971,
-      "grad_norm": 0.04900379851460457,
+      "grad_norm": 0.049401551485061646,
       "learning_rate": 0.00022945544554455445,
-      "loss": 0.9017,
+      "loss": 0.9016,
       "step": 570
     },
     {
       "epoch": 0.7171561051004637,
-      "grad_norm": 0.0502135343849659,
+      "grad_norm": 0.049783241003751755,
       "learning_rate": 0.0002282178217821782,
       "loss": 0.8774,
       "step": 580
     },
     {
       "epoch": 0.7295208655332303,
-      "grad_norm": 0.047438524663448334,
+      "grad_norm": 0.04755168408155441,
       "learning_rate": 0.00022698019801980194,
-      "loss": 0.8815,
+      "loss": 0.8814,
       "step": 590
     },
     {
       "epoch": 0.7418856259659969,
-      "grad_norm": 0.04633660241961479,
+      "grad_norm": 0.04885553568601608,
       "learning_rate": 0.00022574257425742573,
-      "loss": 0.8746,
+      "loss": 0.8744,
       "step": 600
     },
     {
       "epoch": 0.7542503863987635,
-      "grad_norm": 0.04724352806806564,
+      "grad_norm": 0.04771718755364418,
       "learning_rate": 0.0002245049504950495,
-      "loss": 0.9005,
+      "loss": 0.9001,
       "step": 610
     },
     {
       "epoch": 0.7666151468315301,
-      "grad_norm": 0.04557831212878227,
+      "grad_norm": 0.04642605781555176,
       "learning_rate": 0.00022326732673267323,
-      "loss": 0.8798,
+      "loss": 0.88,
       "step": 620
     },
     {
       "epoch": 0.7789799072642968,
-      "grad_norm": 0.047848157584667206,
+      "grad_norm": 0.047350749373435974,
       "learning_rate": 0.00022202970297029702,
-      "loss": 0.893,
+      "loss": 0.8928,
       "step": 630
     },
     {
       "epoch": 0.7913446676970634,
-      "grad_norm": 0.04449377954006195,
+      "grad_norm": 0.04467844218015671,
       "learning_rate": 0.00022079207920792078,
-      "loss": 0.8851,
+      "loss": 0.885,
       "step": 640
     },
     {
       "epoch": 0.80370942812983,
-      "grad_norm": 0.04431360587477684,
+      "grad_norm": 0.04457986727356911,
       "learning_rate": 0.0002195544554455445,
-      "loss": 0.8876,
+      "loss": 0.888,
       "step": 650
     },
     {
       "epoch": 0.8160741885625966,
-      "grad_norm": 0.04388862103223801,
+      "grad_norm": 0.04410697519779205,
       "learning_rate": 0.0002183168316831683,
-      "loss": 0.8886,
+      "loss": 0.8888,
       "step": 660
     },
     {
       "epoch": 0.8284389489953632,
-      "grad_norm": 0.04719037562608719,
+      "grad_norm": 0.0475030243396759,
       "learning_rate": 0.00021707920792079206,
-      "loss": 0.9011,
+      "loss": 0.9009,
       "step": 670
     },
     {
       "epoch": 0.8408037094281299,
-      "grad_norm": 0.04291271045804024,
+      "grad_norm": 0.043028101325035095,
       "learning_rate": 0.00021584158415841585,
-      "loss": 0.8738,
+      "loss": 0.8735,
       "step": 680
     },
     {
       "epoch": 0.8531684698608965,
-      "grad_norm": 0.04412473365664482,
+      "grad_norm": 0.04463913291692734,
       "learning_rate": 0.00021460396039603958,
-      "loss": 0.8738,
+      "loss": 0.8742,
       "step": 690
     },
     {
       "epoch": 0.865533230293663,
-      "grad_norm": 0.046331875026226044,
+      "grad_norm": 0.04648848995566368,
       "learning_rate": 0.00021336633663366334,
       "loss": 0.8899,
       "step": 700
     },
     {
       "epoch": 0.8778979907264297,
-      "grad_norm": 0.04418357461690903,
+      "grad_norm": 0.04463621601462364,
       "learning_rate": 0.00021212871287128713,
-      "loss": 0.889,
+      "loss": 0.8887,
       "step": 710
     },
     {
       "epoch": 0.8902627511591963,
-      "grad_norm": 0.04221678525209427,
+      "grad_norm": 0.04241452366113663,
       "learning_rate": 0.00021089108910891087,
-      "loss": 0.8751,
+      "loss": 0.8749,
       "step": 720
     },
     {
       "epoch": 0.9026275115919629,
-      "grad_norm": 0.04472072795033455,
+      "grad_norm": 0.04464114084839821,
       "learning_rate": 0.00020965346534653463,
-      "loss": 0.8775,
+      "loss": 0.8774,
       "step": 730
     },
     {
       "epoch": 0.9149922720247295,
-      "grad_norm": 0.04348697140812874,
+      "grad_norm": 0.04345027729868889,
       "learning_rate": 0.00020841584158415842,
-      "loss": 0.8751,
+      "loss": 0.8753,
       "step": 740
     },
     {
       "epoch": 0.9273570324574961,
-      "grad_norm": 0.04851846024394035,
+      "grad_norm": 0.048532094806432724,
       "learning_rate": 0.00020717821782178215,
-      "loss": 0.8951,
+      "loss": 0.8946,
       "step": 750
     },
     {
       "epoch": 0.9397217928902627,
-      "grad_norm": 0.04079887643456459,
+      "grad_norm": 0.04126739129424095,
       "learning_rate": 0.0002059405940594059,
-      "loss": 0.9028,
+      "loss": 0.903,
       "step": 760
     },
     {
       "epoch": 0.9520865533230294,
-      "grad_norm": 0.04408387467265129,
+      "grad_norm": 0.04423375427722931,
       "learning_rate": 0.0002047029702970297,
-      "loss": 0.8842,
+      "loss": 0.8843,
       "step": 770
     },
     {
       "epoch": 0.964451313755796,
-      "grad_norm": 0.04127747192978859,
+      "grad_norm": 0.04136930778622627,
       "learning_rate": 0.00020346534653465346,
-      "loss": 0.876,
+      "loss": 0.8757,
       "step": 780
     },
     {
       "epoch": 0.9768160741885626,
-      "grad_norm": 0.05341732129454613,
+      "grad_norm": 0.05331163853406906,
       "learning_rate": 0.0002022277227722772,
-      "loss": 0.8843,
+      "loss": 0.8842,
       "step": 790
     },
     {
       "epoch": 0.9891808346213292,
-      "grad_norm": 0.04694453999400139,
+      "grad_norm": 0.04790889099240303,
       "learning_rate": 0.00020099009900990098,
-      "loss": 0.8815,
+      "loss": 0.8814,
       "step": 800
     },
     {
       "epoch": 1.0012364760432766,
-      "grad_norm": 0.05160349979996681,
+      "grad_norm": 0.05177275091409683,
       "learning_rate": 0.00019975247524752475,
-      "loss": 0.8857,
+      "loss": 0.8858,
       "step": 810
     },
     {
       "epoch": 1.0136012364760432,
-      "grad_norm": 0.0415058434009552,
+      "grad_norm": 0.0411980040371418,
       "learning_rate": 0.00019851485148514848,
       "loss": 0.8461,
       "step": 820
     },
     {
       "epoch": 1.02596599690881,
-      "grad_norm": 0.04516944661736488,
+      "grad_norm": 0.04518349468708038,
       "learning_rate": 0.00019727722772277227,
-      "loss": 0.8548,
+      "loss": 0.8547,
       "step": 830
     },
     {
       "epoch": 1.0383307573415765,
-      "grad_norm": 0.04707406461238861,
+      "grad_norm": 0.047048419713974,
       "learning_rate": 0.00019603960396039603,
-      "loss": 0.8503,
+      "loss": 0.8502,
       "step": 840
     },
     {
       "epoch": 1.0506955177743431,
-      "grad_norm": 0.049354761838912964,
+      "grad_norm": 0.04998902231454849,
       "learning_rate": 0.00019480198019801976,
       "loss": 0.8584,
       "step": 850
     },
     {
       "epoch": 1.0630602782071097,
-      "grad_norm": 0.04959525167942047,
+      "grad_norm": 0.05004483088850975,
       "learning_rate": 0.00019356435643564355,
-      "loss": 0.8788,
+      "loss": 0.8787,
       "step": 860
     },
     {
       "epoch": 1.0754250386398763,
-      "grad_norm": 0.048685044050216675,
+      "grad_norm": 0.0483798012137413,
       "learning_rate": 0.0001923267326732673,
-      "loss": 0.873,
+      "loss": 0.8732,
       "step": 870
     },
     {
       "epoch": 1.087789799072643,
-      "grad_norm": 0.045906998217105865,
+      "grad_norm": 0.048114124685525894,
       "learning_rate": 0.00019108910891089107,
-      "loss": 0.8775,
+      "loss": 0.8773,
       "step": 880
     },
     {
       "epoch": 1.1001545595054096,
-      "grad_norm": 0.04486127197742462,
+      "grad_norm": 0.04553611949086189,
       "learning_rate": 0.00018985148514851484,
-      "loss": 0.8647,
+      "loss": 0.8646,
       "step": 890
     },
     {
       "epoch": 1.1125193199381762,
-      "grad_norm": 0.05211256071925163,
+      "grad_norm": 0.052288319915533066,
       "learning_rate": 0.0001886138613861386,
-      "loss": 0.8594,
+      "loss": 0.8592,
       "step": 900
     },
     {
       "epoch": 1.1248840803709428,
-      "grad_norm": 0.05048747360706329,
+      "grad_norm": 0.05070117861032486,
       "learning_rate": 0.00018737623762376236,
-      "loss": 0.8564,
+      "loss": 0.8565,
       "step": 910
     },
     {
       "epoch": 1.1372488408037094,
-      "grad_norm": 0.04840526729822159,
+      "grad_norm": 0.049008361995220184,
       "learning_rate": 0.00018613861386138612,
-      "loss": 0.8782,
+      "loss": 0.8783,
       "step": 920
     },
     {
       "epoch": 1.1496136012364762,
-      "grad_norm": 0.049940045922994614,
+      "grad_norm": 0.04916449636220932,
       "learning_rate": 0.00018490099009900988,
-      "loss": 0.867,
+      "loss": 0.8668,
       "step": 930
     },
     {
       "epoch": 1.1619783616692427,
-      "grad_norm": 0.053729794919490814,
+      "grad_norm": 0.05646826699376106,
       "learning_rate": 0.00018366336633663364,
-      "loss": 0.8585,
+      "loss": 0.858,
       "step": 940
     },
     {
       "epoch": 1.1743431221020093,
-      "grad_norm": 0.05020948871970177,
+      "grad_norm": 0.05039024353027344,
       "learning_rate": 0.0001824257425742574,
-      "loss": 0.8688,
+      "loss": 0.8687,
       "step": 950
     },
     {
       "epoch": 1.1867078825347759,
-      "grad_norm": 0.0517219677567482,
+      "grad_norm": 0.052257779985666275,
       "learning_rate": 0.00018118811881188116,
       "loss": 0.8731,
       "step": 960
     },
     {
       "epoch": 1.1990726429675425,
-      "grad_norm": 0.04891285300254822,
+      "grad_norm": 0.04960246384143829,
       "learning_rate": 0.00017995049504950493,
-      "loss": 0.8348,
+      "loss": 0.8346,
       "step": 970
     },
     {
       "epoch": 1.211437403400309,
-      "grad_norm": 0.051312196999788284,
+      "grad_norm": 0.05193152651190758,
       "learning_rate": 0.00017871287128712871,
-      "loss": 0.8658,
+      "loss": 0.8656,
       "step": 980
     },
     {
       "epoch": 1.2238021638330758,
-      "grad_norm": 0.051922161132097244,
+      "grad_norm": 0.05180949717760086,
       "learning_rate": 0.00017747524752475245,
       "loss": 0.8542,
       "step": 990
     },
     {
       "epoch": 1.2361669242658424,
-      "grad_norm": 0.0521603561937809,
+      "grad_norm": 0.05225878953933716,
       "learning_rate": 0.0001762376237623762,
       "loss": 0.8628,
       "step": 1000
     },
     {
       "epoch": 1.248531684698609,
-      "grad_norm": 0.05443425104022026,
+      "grad_norm": 0.05485387519001961,
       "learning_rate": 0.000175,
-      "loss": 0.875,
+      "loss": 0.8746,
       "step": 1010
     },
     {
       "epoch": 1.2608964451313756,
-      "grad_norm": 0.05506913363933563,
+      "grad_norm": 0.06754795461893082,
       "learning_rate": 0.00017376237623762373,
-      "loss": 0.8704,
+      "loss": 0.8702,
       "step": 1020
     },
     {
       "epoch": 1.2732612055641421,
-      "grad_norm": 0.05535837262868881,
+      "grad_norm": 0.05525548383593559,
       "learning_rate": 0.00017252475247524752,
-      "loss": 0.8629,
+      "loss": 0.863,
       "step": 1030
     },
     {
       "epoch": 1.2856259659969087,
-      "grad_norm": 0.050953879952430725,
+      "grad_norm": 0.05193280428647995,
       "learning_rate": 0.00017128712871287128,
-      "loss": 0.8386,
+      "loss": 0.8389,
       "step": 1040
     },
     {
       "epoch": 1.2979907264296755,
-      "grad_norm": 0.047925543040037155,
+      "grad_norm": 0.04822159186005592,
       "learning_rate": 0.00017004950495049502,
-      "loss": 0.8664,
+      "loss": 0.8665,
       "step": 1050
     },
     {
       "epoch": 1.310355486862442,
-      "grad_norm": 0.054691240191459656,
+      "grad_norm": 0.05497356876730919,
       "learning_rate": 0.0001688118811881188,
-      "loss": 0.8634,
+      "loss": 0.8635,
       "step": 1060
     },
     {
       "epoch": 1.3227202472952087,
-      "grad_norm": 0.05087495222687721,
+      "grad_norm": 0.05118054896593094,
       "learning_rate": 0.00016757425742574257,
-      "loss": 0.8482,
+      "loss": 0.8483,
       "step": 1070
     },
     {
       "epoch": 1.3350850077279752,
-      "grad_norm": 0.051902078092098236,
+      "grad_norm": 0.051902711391448975,
       "learning_rate": 0.00016633663366336633,
       "loss": 0.8478,
       "step": 1080
     },
     {
       "epoch": 1.3474497681607418,
-      "grad_norm": 0.05033488944172859,
+      "grad_norm": 0.049953706562519073,
       "learning_rate": 0.0001650990099009901,
-      "loss": 0.8572,
+      "loss": 0.8569,
       "step": 1090
     },
     {
       "epoch": 1.3598145285935086,
-      "grad_norm": 0.05153260752558708,
+      "grad_norm": 0.09028486907482147,
       "learning_rate": 0.00016386138613861385,
       "loss": 0.8465,
       "step": 1100
     },
     {
       "epoch": 1.3721792890262752,
-      "grad_norm": 0.052806247025728226,
+      "grad_norm": 0.05248475819826126,
       "learning_rate": 0.0001626237623762376,
       "loss": 0.8707,
       "step": 1110
     },
     {
       "epoch": 1.3845440494590417,
-      "grad_norm": 0.05425600707530975,
+      "grad_norm": 0.05470622703433037,
       "learning_rate": 0.00016138613861386137,
-      "loss": 0.858,
+      "loss": 0.8581,
       "step": 1120
     },
     {
       "epoch": 1.3969088098918083,
-      "grad_norm": 0.05116913095116615,
+      "grad_norm": 0.051429346203804016,
       "learning_rate": 0.00016014851485148513,
       "loss": 0.867,
       "step": 1130
     },
     {
       "epoch": 1.409273570324575,
-      "grad_norm": 0.052799541503190994,
+      "grad_norm": 0.05353890359401703,
       "learning_rate": 0.0001589108910891089,
-      "loss": 0.849,
+      "loss": 0.8489,
       "step": 1140
     },
     {
       "epoch": 1.4216383307573417,
-      "grad_norm": 0.06275513023138046,
+      "grad_norm": 0.0630929172039032,
       "learning_rate": 0.00015767326732673266,
-      "loss": 0.8577,
+      "loss": 0.8575,
       "step": 1150
     },
     {
       "epoch": 1.4340030911901083,
-      "grad_norm": 0.051965054124593735,
+      "grad_norm": 0.0524783730506897,
       "learning_rate": 0.00015643564356435642,
-      "loss": 0.853,
+      "loss": 0.8527,
       "step": 1160
     },
     {
       "epoch": 1.4463678516228748,
-      "grad_norm": 0.05356529727578163,
+      "grad_norm": 0.05413209646940231,
       "learning_rate": 0.00015519801980198018,
-      "loss": 0.8789,
+      "loss": 0.8786,
       "step": 1170
     },
     {
       "epoch": 1.4587326120556414,
-      "grad_norm": 0.05566537007689476,
+      "grad_norm": 0.055751536041498184,
       "learning_rate": 0.00015396039603960397,
-      "loss": 0.8716,
+      "loss": 0.872,
       "step": 1180
     },
     {
       "epoch": 1.471097372488408,
-      "grad_norm": 0.05320986732840538,
+      "grad_norm": 0.05271457880735397,
       "learning_rate": 0.0001527227722772277,
-      "loss": 0.8736,
+      "loss": 0.8734,
       "step": 1190
     },
     {
       "epoch": 1.4834621329211746,
-      "grad_norm": 0.049232013523578644,
+      "grad_norm": 0.04827325418591499,
       "learning_rate": 0.00015148514851485146,
-      "loss": 0.849,
+      "loss": 0.8488,
       "step": 1200
     },
     {
       "epoch": 1.4958268933539411,
-      "grad_norm": 0.058629848062992096,
+      "grad_norm": 0.05717690661549568,
       "learning_rate": 0.00015024752475247525,
       "loss": 0.8732,
       "step": 1210
     },
     {
       "epoch": 1.508191653786708,
-      "grad_norm": 0.055390194058418274,
+      "grad_norm": 0.055509038269519806,
       "learning_rate": 0.000149009900990099,
-      "loss": 0.8678,
+      "loss": 0.8675,
       "step": 1220
     },
     {
       "epoch": 1.5205564142194745,
-      "grad_norm": 0.05527270585298538,
+      "grad_norm": 0.05562078580260277,
       "learning_rate": 0.00014777227722772275,
-      "loss": 0.8643,
+      "loss": 0.8644,
       "step": 1230
     },
     {
       "epoch": 1.532921174652241,
-      "grad_norm": 0.04652067646384239,
+      "grad_norm": 0.046674925833940506,
       "learning_rate": 0.00014653465346534653,
       "loss": 0.8429,
       "step": 1240
     },
     {
       "epoch": 1.545285935085008,
-      "grad_norm": 0.05379781499505043,
+      "grad_norm": 0.053251732140779495,
       "learning_rate": 0.0001452970297029703,
-      "loss": 0.8492,
+      "loss": 0.849,
       "step": 1250
     },
     {
       "epoch": 1.5576506955177742,
-      "grad_norm": 0.05249844118952751,
+      "grad_norm": 0.05253510922193527,
       "learning_rate": 0.00014405940594059403,
-      "loss": 0.8447,
+      "loss": 0.8445,
       "step": 1260
     },
     {
       "epoch": 1.570015455950541,
-      "grad_norm": 0.049639806151390076,
+      "grad_norm": 0.05021601915359497,
       "learning_rate": 0.00014282178217821782,
-      "loss": 0.8667,
+      "loss": 0.8668,
       "step": 1270
     },
     {
       "epoch": 1.5823802163833076,
-      "grad_norm": 0.052679967135190964,
+      "grad_norm": 0.052446555346250534,
       "learning_rate": 0.00014158415841584158,
-      "loss": 0.8739,
+      "loss": 0.8733,
       "step": 1280
     },
     {
       "epoch": 1.5947449768160742,
-      "grad_norm": 0.05990573391318321,
+      "grad_norm": 0.056364450603723526,
       "learning_rate": 0.00014034653465346534,
-      "loss": 0.8824,
+      "loss": 0.8823,
       "step": 1290
     },
     {
       "epoch": 1.6071097372488408,
-      "grad_norm": 0.052240803837776184,
+      "grad_norm": 0.05288272723555565,
       "learning_rate": 0.0001391089108910891,
-      "loss": 0.868,
+      "loss": 0.8678,
       "step": 1300
     },
     {
       "epoch": 1.6194744976816073,
-      "grad_norm": 0.05380776897072792,
+      "grad_norm": 0.054042939096689224,
       "learning_rate": 0.00013787128712871286,
-      "loss": 0.8442,
+      "loss": 0.8439,
       "step": 1310
     },
     {
       "epoch": 1.6318392581143741,
-      "grad_norm": 0.05081896856427193,
+      "grad_norm": 0.051554158329963684,
       "learning_rate": 0.00013663366336633662,
-      "loss": 0.8515,
+      "loss": 0.8514,
       "step": 1320
     },
     {
       "epoch": 1.6442040185471405,
-      "grad_norm": 0.04869316518306732,
+      "grad_norm": 0.04892382398247719,
       "learning_rate": 0.00013539603960396039,
-      "loss": 0.8339,
+      "loss": 0.834,
       "step": 1330
     },
     {
       "epoch": 1.6565687789799073,
-      "grad_norm": 0.056119490414857864,
+      "grad_norm": 0.05448554828763008,
       "learning_rate": 0.00013415841584158415,
-      "loss": 0.8483,
+      "loss": 0.8484,
       "step": 1340
     },
     {
       "epoch": 1.6689335394126739,
-      "grad_norm": 0.05742491036653519,
+      "grad_norm": 0.056680306792259216,
       "learning_rate": 0.0001329207920792079,
-      "loss": 0.8494,
+      "loss": 0.8495,
       "step": 1350
     },
     {
       "epoch": 1.6812982998454404,
-      "grad_norm": 0.055017951875925064,
+      "grad_norm": 0.05566761642694473,
       "learning_rate": 0.00013168316831683167,
-      "loss": 0.8563,
+      "loss": 0.856,
       "step": 1360
     },
     {
       "epoch": 1.6936630602782072,
-      "grad_norm": 0.04963842034339905,
+      "grad_norm": 0.04952670633792877,
       "learning_rate": 0.00013044554455445543,
       "loss": 0.8405,
       "step": 1370
     },
     {
       "epoch": 1.7060278207109736,
-      "grad_norm": 0.05574873462319374,
+      "grad_norm": 0.05578543245792389,
       "learning_rate": 0.0001292079207920792,
-      "loss": 0.8557,
+      "loss": 0.8555,
       "step": 1380
     },
     {
       "epoch": 1.7183925811437404,
-      "grad_norm": 0.05482814088463783,
+      "grad_norm": 0.05533617362380028,
       "learning_rate": 0.00012797029702970295,
-      "loss": 0.8559,
+      "loss": 0.8558,
       "step": 1390
     },
     {
       "epoch": 1.730757341576507,
-      "grad_norm": 0.06040499359369278,
+      "grad_norm": 0.05991559103131294,
       "learning_rate": 0.00012673267326732672,
-      "loss": 0.8638,
+      "loss": 0.8636,
       "step": 1400
     },
     {
       "epoch": 1.7431221020092735,
-      "grad_norm": 0.05430367961525917,
+      "grad_norm": 0.054518427699804306,
       "learning_rate": 0.00012549504950495048,
-      "loss": 0.8473,
+      "loss": 0.8471,
       "step": 1410
     },
     {
       "epoch": 1.7554868624420403,
-      "grad_norm": 0.048315104097127914,
+      "grad_norm": 0.04764275252819061,
       "learning_rate": 0.00012425742574257426,
-      "loss": 0.845,
+      "loss": 0.8449,
       "step": 1420
     },
     {
       "epoch": 1.7678516228748067,
-      "grad_norm": 0.05943458899855614,
+      "grad_norm": 0.058475952595472336,
       "learning_rate": 0.000123019801980198,
-      "loss": 0.853,
+      "loss": 0.8523,
       "step": 1430
     },
     {
       "epoch": 1.7802163833075735,
-      "grad_norm": 0.05744357407093048,
+      "grad_norm": 0.05991446226835251,
       "learning_rate": 0.00012178217821782177,
-      "loss": 0.8562,
+      "loss": 0.8564,
       "step": 1440
     },
     {
       "epoch": 1.79258114374034,
-      "grad_norm": 0.06155743822455406,
+      "grad_norm": 0.0623490996658802,
       "learning_rate": 0.00012054455445544554,
       "loss": 0.8404,
       "step": 1450
     },
     {
       "epoch": 1.8049459041731066,
-      "grad_norm": 0.04887942224740982,
+      "grad_norm": 0.04905753955245018,
       "learning_rate": 0.0001193069306930693,
-      "loss": 0.8477,
+      "loss": 0.8474,
       "step": 1460
     },
     {
       "epoch": 1.8173106646058734,
-      "grad_norm": 0.05377992242574692,
+      "grad_norm": 0.05426807701587677,
       "learning_rate": 0.00011806930693069306,
-      "loss": 0.8765,
+      "loss": 0.8766,
       "step": 1470
     },
     {
       "epoch": 1.8296754250386398,
-      "grad_norm": 0.0468844473361969,
+      "grad_norm": 0.0476132333278656,
       "learning_rate": 0.00011683168316831682,
-      "loss": 0.8151,
+      "loss": 0.815,
       "step": 1480
     },
     {
       "epoch": 1.8420401854714066,
-      "grad_norm": 0.05763052776455879,
+      "grad_norm": 0.05849111080169678,
       "learning_rate": 0.0001155940594059406,
-      "loss": 0.8539,
+      "loss": 0.854,
       "step": 1490
     },
     {
       "epoch": 1.8544049459041732,
-      "grad_norm": 0.054946307092905045,
+      "grad_norm": 0.05493124946951866,
       "learning_rate": 0.00011435643564356434,
-      "loss": 0.8564,
+      "loss": 0.8563,
       "step": 1500
     },
     {
       "epoch": 1.8667697063369397,
-      "grad_norm": 0.060760248452425,
+      "grad_norm": 0.05999801307916641,
       "learning_rate": 0.0001131188118811881,
       "loss": 0.8498,
       "step": 1510
     },
     {
       "epoch": 1.8791344667697063,
-      "grad_norm": 0.058039598166942596,
+      "grad_norm": 0.058151423931121826,
       "learning_rate": 0.00011188118811881188,
-      "loss": 0.8647,
+      "loss": 0.8645,
       "step": 1520
     },
     {
       "epoch": 1.8914992272024729,
-      "grad_norm": 0.05479070916771889,
+      "grad_norm": 0.05524227395653725,
       "learning_rate": 0.00011064356435643564,
-      "loss": 0.8625,
+      "loss": 0.8624,
       "step": 1530
     },
     {
       "epoch": 1.9038639876352397,
-      "grad_norm": 0.0583939254283905,
+      "grad_norm": 0.06369632482528687,
       "learning_rate": 0.00010940594059405939,
       "loss": 0.8695,
       "step": 1540
     },
     {
       "epoch": 1.916228748068006,
-      "grad_norm": 0.058852337300777435,
+      "grad_norm": 0.057092998176813126,
       "learning_rate": 0.00010816831683168316,
-      "loss": 0.8443,
+      "loss": 0.844,
       "step": 1550
     },
     {
       "epoch": 1.9285935085007728,
-      "grad_norm": 0.05506705492734909,
+      "grad_norm": 0.05554778128862381,
       "learning_rate": 0.00010693069306930692,
-      "loss": 0.8544,
+      "loss": 0.8543,
       "step": 1560
     },
     {
       "epoch": 1.9409582689335394,
-      "grad_norm": 0.05682089179754257,
+      "grad_norm": 0.05691225454211235,
       "learning_rate": 0.00010569306930693068,
-      "loss": 0.8718,
+      "loss": 0.8714,
       "step": 1570
     },
     {
       "epoch": 1.953323029366306,
-      "grad_norm": 0.05604562535881996,
+      "grad_norm": 0.0564524307847023,
       "learning_rate": 0.00010445544554455445,
-      "loss": 0.857,
+      "loss": 0.8574,
       "step": 1580
     },
     {
       "epoch": 1.9656877897990728,
-      "grad_norm": 0.058413226157426834,
+      "grad_norm": 0.0588836595416069,
       "learning_rate": 0.0001032178217821782,
       "loss": 0.8558,
       "step": 1590
     },
     {
       "epoch": 1.9780525502318391,
-      "grad_norm": 0.054590627551078796,
+      "grad_norm": 0.05634515732526779,
       "learning_rate": 0.00010198019801980197,
-      "loss": 0.8443,
+      "loss": 0.8444,
       "step": 1600
     },
     {
       "epoch": 1.990417310664606,
-      "grad_norm": 0.05447821691632271,
+      "grad_norm": 0.055482737720012665,
       "learning_rate": 0.00010074257425742573,
       "loss": 0.8672,
       "step": 1610
     },
     {
       "epoch": 2.002472952086553,
-      "grad_norm": 0.05398769676685333,
+      "grad_norm": 0.054257094860076904,
       "learning_rate": 9.95049504950495e-05,
-      "loss": 0.8582,
+      "loss": 0.8579,
       "step": 1620
     },
     {
       "epoch": 2.01483771251932,
-      "grad_norm": 0.057375263422727585,
+      "grad_norm": 0.05709832161664963,
       "learning_rate": 9.826732673267325e-05,
-      "loss": 0.8418,
+      "loss": 0.8414,
       "step": 1630
     },
     {
       "epoch": 2.0272024729520863,
-      "grad_norm": 0.054974183440208435,
+      "grad_norm": 0.05785168707370758,
       "learning_rate": 9.702970297029701e-05,
-      "loss": 0.8224,
+      "loss": 0.8223,
       "step": 1640
     },
     {
       "epoch": 2.039567233384853,
-      "grad_norm": 0.06044444069266319,
+      "grad_norm": 0.060052480548620224,
       "learning_rate": 9.579207920792079e-05,
-      "loss": 0.8373,
+      "loss": 0.8371,
       "step": 1650
     },
     {
       "epoch": 2.05193199381762,
-      "grad_norm": 0.06379813700914383,
+      "grad_norm": 0.06388446688652039,
       "learning_rate": 9.455445544554454e-05,
-      "loss": 0.8311,
+      "loss": 0.8308,
       "step": 1660
     },
     {
       "epoch": 2.0642967542503863,
-      "grad_norm": 0.05604099482297897,
+      "grad_norm": 0.05495399236679077,
       "learning_rate": 9.331683168316831e-05,
-      "loss": 0.8586,
+      "loss": 0.8582,
       "step": 1670
     },
     {
       "epoch": 2.076661514683153,
-      "grad_norm": 0.05408864468336105,
+      "grad_norm": 0.0544477179646492,
       "learning_rate": 9.207920792079207e-05,
-      "loss": 0.8385,
+      "loss": 0.8383,
       "step": 1680
     },
     {
       "epoch": 2.0890262751159194,
-      "grad_norm": 0.06171610206365585,
+      "grad_norm": 0.06450890749692917,
       "learning_rate": 9.084158415841582e-05,
-      "loss": 0.836,
+      "loss": 0.8359,
       "step": 1690
     },
     {
       "epoch": 2.1013910355486862,
-      "grad_norm": 0.05357811599969864,
+      "grad_norm": 0.054119642823934555,
       "learning_rate": 8.96039603960396e-05,
-      "loss": 0.8365,
+      "loss": 0.8363,
       "step": 1700
     },
     {
       "epoch": 2.113755795981453,
-      "grad_norm": 0.059701114892959595,
+      "grad_norm": 0.05726737529039383,
       "learning_rate": 8.836633663366336e-05,
-      "loss": 0.8168,
+      "loss": 0.8169,
       "step": 1710
     },
     {
       "epoch": 2.1261205564142194,
-      "grad_norm": 0.05693197622895241,
+      "grad_norm": 0.0577755868434906,
       "learning_rate": 8.712871287128713e-05,
-      "loss": 0.8588,
+      "loss": 0.8586,
       "step": 1720
     },
     {
       "epoch": 2.138485316846986,
-      "grad_norm": 0.06465724855661392,
+      "grad_norm": 0.06451012194156647,
       "learning_rate": 8.589108910891088e-05,
-      "loss": 0.8342,
+      "loss": 0.834,
       "step": 1730
     },
     {
       "epoch": 2.1508500772797525,
-      "grad_norm": 0.06339121609926224,
+      "grad_norm": 0.06303463876247406,
       "learning_rate": 8.465346534653464e-05,
-      "loss": 0.8338,
+      "loss": 0.8333,
       "step": 1740
     },
     {
       "epoch": 2.1632148377125193,
-      "grad_norm": 0.05768771097064018,
+      "grad_norm": 0.058561887592077255,
       "learning_rate": 8.341584158415841e-05,
       "loss": 0.8321,
       "step": 1750
     },
     {
       "epoch": 2.175579598145286,
-      "grad_norm": 0.05351224169135094,
+      "grad_norm": 0.05364146828651428,
       "learning_rate": 8.217821782178216e-05,
-      "loss": 0.8426,
+      "loss": 0.8428,
       "step": 1760
     },
     {
       "epoch": 2.1879443585780525,
-      "grad_norm": 0.06381036341190338,
+      "grad_norm": 0.063669353723526,
       "learning_rate": 8.094059405940594e-05,
-      "loss": 0.8531,
+      "loss": 0.8527,
       "step": 1770
     },
     {
       "epoch": 2.2003091190108193,
-      "grad_norm": 0.057617682963609695,
+      "grad_norm": 0.05790480971336365,
       "learning_rate": 7.97029702970297e-05,
-      "loss": 0.8263,
+      "loss": 0.8261,
       "step": 1780
     },
     {
       "epoch": 2.2126738794435856,
-      "grad_norm": 0.06280315667390823,
+      "grad_norm": 0.06101266294717789,
       "learning_rate": 7.846534653465345e-05,
-      "loss": 0.8073,
+      "loss": 0.8075,
       "step": 1790
     },
     {
       "epoch": 2.2250386398763524,
-      "grad_norm": 0.06251993030309677,
+      "grad_norm": 0.06296826899051666,
       "learning_rate": 7.722772277227722e-05,
-      "loss": 0.8285,
+      "loss": 0.8284,
       "step": 1800
     },
     {
       "epoch": 2.237403400309119,
-      "grad_norm": 0.05487222224473953,
+      "grad_norm": 0.0548894926905632,
       "learning_rate": 7.599009900990098e-05,
-      "loss": 0.8389,
+      "loss": 0.8385,
       "step": 1810
     },
     {
       "epoch": 2.2497681607418856,
-      "grad_norm": 0.06212658807635307,
+      "grad_norm": 0.06245751306414604,
       "learning_rate": 7.475247524752474e-05,
-      "loss": 0.819,
+      "loss": 0.8188,
       "step": 1820
     },
     {
       "epoch": 2.2621329211746524,
-      "grad_norm": 0.06791824847459793,
+      "grad_norm": 0.06896353513002396,
       "learning_rate": 7.35148514851485e-05,
-      "loss": 0.823,
+      "loss": 0.8229,
       "step": 1830
     },
     {
       "epoch": 2.2744976816074187,
-      "grad_norm": 0.06564588844776154,
+      "grad_norm": 0.06569264829158783,
       "learning_rate": 7.227722772277227e-05,
-      "loss": 0.8399,
+      "loss": 0.8398,
       "step": 1840
     },
     {
       "epoch": 2.2868624420401855,
-      "grad_norm": 0.07918984442949295,
+      "grad_norm": 0.06732139736413956,
       "learning_rate": 7.103960396039604e-05,
-      "loss": 0.8441,
+      "loss": 0.8439,
       "step": 1850
     },
     {
       "epoch": 2.2992272024729523,
-      "grad_norm": 0.06684021651744843,
+      "grad_norm": 0.06835715472698212,
       "learning_rate": 6.98019801980198e-05,
-      "loss": 0.8213,
+      "loss": 0.821,
       "step": 1860
     },
     {
       "epoch": 2.3115919629057187,
-      "grad_norm": 0.05864300578832626,
+      "grad_norm": 0.05850212648510933,
       "learning_rate": 6.856435643564355e-05,
-      "loss": 0.8238,
+      "loss": 0.8235,
       "step": 1870
     },
     {
       "epoch": 2.3239567233384855,
-      "grad_norm": 0.05827944353222847,
+      "grad_norm": 0.06048553064465523,
       "learning_rate": 6.732673267326732e-05,
-      "loss": 0.8438,
+      "loss": 0.844,
       "step": 1880
     },
     {
       "epoch": 2.336321483771252,
-      "grad_norm": 0.05539786070585251,
+      "grad_norm": 0.05443299934267998,
       "learning_rate": 6.608910891089109e-05,
       "loss": 0.8173,
       "step": 1890
     },
     {
       "epoch": 2.3486862442040186,
-      "grad_norm": 0.06571885198354721,
+      "grad_norm": 0.06576599180698395,
       "learning_rate": 6.485148514851485e-05,
-      "loss": 0.8262,
+      "loss": 0.826,
       "step": 1900
     },
     {
       "epoch": 2.361051004636785,
-      "grad_norm": 0.06220625340938568,
+      "grad_norm": 0.06261160224676132,
       "learning_rate": 6.361386138613861e-05,
-      "loss": 0.8576,
+      "loss": 0.8571,
       "step": 1910
     },
     {
       "epoch": 2.3734157650695518,
-      "grad_norm": 0.0579352080821991,
+      "grad_norm": 0.05812652036547661,
       "learning_rate": 6.237623762376237e-05,
       "loss": 0.8227,
       "step": 1920
     },
     {
       "epoch": 2.3857805255023186,
-      "grad_norm": 0.06193961948156357,
+      "grad_norm": 0.06309802830219269,
       "learning_rate": 6.113861386138613e-05,
-      "loss": 0.8414,
+      "loss": 0.8412,
       "step": 1930
     },
     {
       "epoch": 2.398145285935085,
-      "grad_norm": 0.061364226043224335,
+      "grad_norm": 0.06207476556301117,
       "learning_rate": 5.99009900990099e-05,
-      "loss": 0.8387,
+      "loss": 0.8386,
       "step": 1940
     },
     {
       "epoch": 2.4105100463678517,
-      "grad_norm": 0.05785266309976578,
+      "grad_norm": 0.05841566249728203,
       "learning_rate": 5.866336633663366e-05,
-      "loss": 0.8284,
+      "loss": 0.828,
       "step": 1950
     },
     {
       "epoch": 2.422874806800618,
-      "grad_norm": 0.057832520455121994,
+      "grad_norm": 0.05857423320412636,
       "learning_rate": 5.742574257425742e-05,
-      "loss": 0.8197,
+      "loss": 0.8198,
       "step": 1960
     },
     {
       "epoch": 2.435239567233385,
-      "grad_norm": 0.06421726942062378,
+      "grad_norm": 0.06476933509111404,
       "learning_rate": 5.618811881188118e-05,
-      "loss": 0.8402,
+      "loss": 0.84,
       "step": 1970
     },
     {
       "epoch": 2.4476043276661517,
-      "grad_norm": 0.06815137714147568,
+      "grad_norm": 0.06856492906808853,
       "learning_rate": 5.4950495049504944e-05,
-      "loss": 0.8389,
+      "loss": 0.8386,
       "step": 1980
     },
     {
       "epoch": 2.459969088098918,
-      "grad_norm": 0.06730205565690994,
+      "grad_norm": 0.0675152987241745,
       "learning_rate": 5.371287128712871e-05,
-      "loss": 0.8604,
+      "loss": 0.8603,
       "step": 1990
     },
     {
       "epoch": 2.472333848531685,
-      "grad_norm": 0.05876993387937546,
+      "grad_norm": 0.059057943522930145,
       "learning_rate": 5.247524752475247e-05,
       "loss": 0.8254,
       "step": 2000
     },
     {
       "epoch": 2.484698608964451,
-      "grad_norm": 0.06757384538650513,
+      "grad_norm": 0.06778612732887268,
       "learning_rate": 5.1237623762376234e-05,
-      "loss": 0.8292,
+      "loss": 0.829,
       "step": 2010
     },
     {
       "epoch": 2.497063369397218,
-      "grad_norm": 0.06531625986099243,
+      "grad_norm": 0.0652635246515274,
       "learning_rate": 4.9999999999999996e-05,
       "loss": 0.8321,
       "step": 2020
     },
     {
       "epoch": 2.5094281298299848,
-      "grad_norm": 0.060086678713560104,
+      "grad_norm": 0.0605316124856472,
       "learning_rate": 4.876237623762376e-05,
-      "loss": 0.8459,
+      "loss": 0.8458,
       "step": 2030
     },
     {
       "epoch": 2.521792890262751,
-      "grad_norm": 0.06336929649114609,
+      "grad_norm": 0.06351178884506226,
       "learning_rate": 4.752475247524752e-05,
-      "loss": 0.82,
+      "loss": 0.8199,
       "step": 2040
     },
     {
       "epoch": 2.534157650695518,
-      "grad_norm": 0.06393607705831528,
+      "grad_norm": 0.0644257590174675,
       "learning_rate": 4.6287128712871286e-05,
-      "loss": 0.8313,
+      "loss": 0.8311,
       "step": 2050
     },
     {
       "epoch": 2.5465224111282843,
-      "grad_norm": 0.06480514258146286,
+      "grad_norm": 0.06502491235733032,
       "learning_rate": 4.504950495049505e-05,
-      "loss": 0.8442,
+      "loss": 0.8443,
       "step": 2060
     },
     {
       "epoch": 2.558887171561051,
-      "grad_norm": 0.07233238965272903,
+      "grad_norm": 0.07183568179607391,
       "learning_rate": 4.38118811881188e-05,
-      "loss": 0.8256,
+      "loss": 0.825,
       "step": 2070
     },
     {
       "epoch": 2.5712519319938174,
-      "grad_norm": 0.06636520475149155,
+      "grad_norm": 0.06714395433664322,
       "learning_rate": 4.257425742574257e-05,
-      "loss": 0.8284,
+      "loss": 0.8283,
       "step": 2080
     },
     {
       "epoch": 2.583616692426584,
-      "grad_norm": 0.07016933709383011,
+      "grad_norm": 0.07098986953496933,
       "learning_rate": 4.133663366336633e-05,
-      "loss": 0.8378,
+      "loss": 0.8376,
       "step": 2090
     },
     {
       "epoch": 2.595981452859351,
-      "grad_norm": 0.06668656319379807,
+      "grad_norm": 0.0671941488981247,
       "learning_rate": 4.00990099009901e-05,
-      "loss": 0.8461,
+      "loss": 0.8457,
       "step": 2100
     },
     {
       "epoch": 2.6083462132921174,
-      "grad_norm": 0.07296927273273468,
+      "grad_norm": 0.07306034862995148,
       "learning_rate": 3.886138613861386e-05,
-      "loss": 0.8432,
+      "loss": 0.843,
       "step": 2110
     },
     {
       "epoch": 2.620710973724884,
-      "grad_norm": 0.06670273840427399,
+      "grad_norm": 0.06762495636940002,
       "learning_rate": 3.7623762376237615e-05,
-      "loss": 0.819,
+      "loss": 0.8188,
       "step": 2120
     },
     {
       "epoch": 2.633075734157651,
-      "grad_norm": 0.060203880071640015,
+      "grad_norm": 0.06061069294810295,
       "learning_rate": 3.638613861386138e-05,
-      "loss": 0.8063,
+      "loss": 0.8059,
       "step": 2130
     },
     {
       "epoch": 2.6454404945904173,
-      "grad_norm": 0.06635984778404236,
+      "grad_norm": 0.0667000338435173,
       "learning_rate": 3.5148514851485144e-05,
       "loss": 0.8364,
       "step": 2140
     },
     {
       "epoch": 2.6578052550231837,
-      "grad_norm": 0.060412149876356125,
+      "grad_norm": 0.058926161378622055,
       "learning_rate": 3.3910891089108906e-05,
-      "loss": 0.827,
+      "loss": 0.8267,
       "step": 2150
     },
     {
       "epoch": 2.6701700154559505,
-      "grad_norm": 0.05948295816779137,
+      "grad_norm": 0.05975179746747017,
       "learning_rate": 3.267326732673267e-05,
-      "loss": 0.802,
+      "loss": 0.8018,
       "step": 2160
     },
     {
       "epoch": 2.6825347758887172,
-      "grad_norm": 0.06251130253076553,
+      "grad_norm": 0.06300190091133118,
       "learning_rate": 3.1435643564356435e-05,
-      "loss": 0.8302,
+      "loss": 0.8306,
       "step": 2170
     },
     {
       "epoch": 2.6948995363214836,
-      "grad_norm": 0.06650058180093765,
+      "grad_norm": 0.06579259783029556,
       "learning_rate": 3.0198019801980193e-05,
-      "loss": 0.8386,
+      "loss": 0.8385,
       "step": 2180
     },
     {
       "epoch": 2.7072642967542504,
-      "grad_norm": 0.07029715925455093,
+      "grad_norm": 0.07062911242246628,
       "learning_rate": 2.8960396039603958e-05,
-      "loss": 0.8421,
+      "loss": 0.8422,
       "step": 2190
     },
     {
       "epoch": 2.719629057187017,
-      "grad_norm": 0.06135771796107292,
+      "grad_norm": 0.06216396763920784,
       "learning_rate": 2.772277227722772e-05,
-      "loss": 0.829,
+      "loss": 0.8292,
       "step": 2200
     },
     {
       "epoch": 2.7319938176197835,
-      "grad_norm": 0.06303984671831131,
+      "grad_norm": 0.06445206701755524,
       "learning_rate": 2.6485148514851484e-05,
-      "loss": 0.8408,
+      "loss": 0.8406,
       "step": 2210
     },
     {
       "epoch": 2.7443585780525503,
-      "grad_norm": 0.06396885961294174,
+      "grad_norm": 0.06448670476675034,
       "learning_rate": 2.5247524752475248e-05,
-      "loss": 0.817,
+      "loss": 0.8171,
       "step": 2220
     },
     {
       "epoch": 2.7567233384853167,
-      "grad_norm": 0.05814013257622719,
+      "grad_norm": 0.05858496576547623,
       "learning_rate": 2.4009900990099006e-05,
-      "loss": 0.8384,
+      "loss": 0.8383,
       "step": 2230
     },
     {
       "epoch": 2.7690880989180835,
-      "grad_norm": 0.07185972481966019,
+      "grad_norm": 0.07208121567964554,
       "learning_rate": 2.277227722772277e-05,
-      "loss": 0.8143,
+      "loss": 0.8145,
       "step": 2240
     },
     {
       "epoch": 2.78145285935085,
-      "grad_norm": 0.06624460965394974,
+      "grad_norm": 0.06663426011800766,
       "learning_rate": 2.1534653465346532e-05,
-      "loss": 0.8296,
+      "loss": 0.8293,
       "step": 2250
     },
     {
       "epoch": 2.7938176197836166,
-      "grad_norm": 0.06510159373283386,
+      "grad_norm": 0.06585463881492615,
       "learning_rate": 2.0297029702970297e-05,
-      "loss": 0.8302,
+      "loss": 0.8303,
       "step": 2260
     },
     {
       "epoch": 2.8061823802163834,
-      "grad_norm": 0.06376007944345474,
+      "grad_norm": 0.06423688679933548,
       "learning_rate": 1.9059405940594058e-05,
-      "loss": 0.8173,
+      "loss": 0.8172,
       "step": 2270
     },
     {
       "epoch": 2.81854714064915,
-      "grad_norm": 0.0644875094294548,
+      "grad_norm": 0.06450697034597397,
       "learning_rate": 1.782178217821782e-05,
-      "loss": 0.8358,
+      "loss": 0.8355,
       "step": 2280
     },
     {
       "epoch": 2.8309119010819166,
-      "grad_norm": 0.05571649968624115,
+      "grad_norm": 0.05580071732401848,
       "learning_rate": 1.6584158415841584e-05,
-      "loss": 0.8139,
+      "loss": 0.8136,
       "step": 2290
     },
     {
       "epoch": 2.8432766615146834,
-      "grad_norm": 0.06556589901447296,
+      "grad_norm": 0.06626173853874207,
       "learning_rate": 1.5346534653465345e-05,
-      "loss": 0.8241,
+      "loss": 0.8238,
       "step": 2300
     },
     {
       "epoch": 2.8556414219474497,
-      "grad_norm": 0.06124770641326904,
+      "grad_norm": 0.061952993273735046,
       "learning_rate": 1.4108910891089108e-05,
-      "loss": 0.8184,
+      "loss": 0.8179,
       "step": 2310
     },
     {
       "epoch": 2.8680061823802165,
-      "grad_norm": 0.06707081943750381,
+      "grad_norm": 0.07288029789924622,
       "learning_rate": 1.287128712871287e-05,
       "loss": 0.8194,
       "step": 2320
     },
     {
       "epoch": 2.880370942812983,
-      "grad_norm": 0.06600210070610046,
+      "grad_norm": 0.06706374138593674,
       "learning_rate": 1.1633663366336632e-05,
-      "loss": 0.8614,
+      "loss": 0.8611,
       "step": 2330
     },
     {
       "epoch": 2.8927357032457497,
-      "grad_norm": 0.06310021132230759,
+      "grad_norm": 0.06370951235294342,
       "learning_rate": 1.0396039603960395e-05,
-      "loss": 0.8485,
+      "loss": 0.848,
       "step": 2340
     },
     {
       "epoch": 2.905100463678516,
-      "grad_norm": 0.06110014021396637,
+      "grad_norm": 0.061200667172670364,
       "learning_rate": 9.158415841584158e-06,
-      "loss": 0.8171,
+      "loss": 0.817,
       "step": 2350
     },
     {
       "epoch": 2.917465224111283,
-      "grad_norm": 0.07767624408006668,
+      "grad_norm": 0.07799932360649109,
       "learning_rate": 7.92079207920792e-06,
-      "loss": 0.8403,
+      "loss": 0.8401,
       "step": 2360
     },
     {
       "epoch": 2.9298299845440496,
-      "grad_norm": 0.07363846898078918,
+      "grad_norm": 0.0740487277507782,
       "learning_rate": 6.683168316831683e-06,
-      "loss": 0.8453,
+      "loss": 0.8447,
       "step": 2370
     },
     {
       "epoch": 2.942194744976816,
-      "grad_norm": 0.06242545694112778,
+      "grad_norm": 0.062499478459358215,
       "learning_rate": 5.445544554455446e-06,
-      "loss": 0.8448,
+      "loss": 0.8446,
       "step": 2380
     },
     {
       "epoch": 2.954559505409583,
-      "grad_norm": 0.06873054057359695,
+      "grad_norm": 0.06899666786193848,
       "learning_rate": 4.207920792079208e-06,
-      "loss": 0.8412,
+      "loss": 0.8411,
       "step": 2390
     },
     {
       "epoch": 2.966924265842349,
-      "grad_norm": 0.06264466792345047,
+      "grad_norm": 0.0634492039680481,
       "learning_rate": 2.97029702970297e-06,
-      "loss": 0.8426,
+      "loss": 0.8422,
       "step": 2400
     },
     {
       "epoch": 2.979289026275116,
-      "grad_norm": 0.06905832886695862,
+      "grad_norm": 0.0700407549738884,
       "learning_rate": 1.7326732673267324e-06,
-      "loss": 0.8469,
+      "loss": 0.8468,
       "step": 2410
     },
     {
       "epoch": 2.9916537867078823,
-      "grad_norm": 0.061685774475336075,
+      "grad_norm": 0.061774324625730515,
       "learning_rate": 4.95049504950495e-07,
-      "loss": 0.8339,
+      "loss": 0.8337,
       "step": 2420
     }
   ],
@@ -1720,7 +1720,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.723183421298311e+18,
+  "total_flos": 2.723181741683245e+18,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:916d7fb58500db212016cbc23cab8f6e9cc2c103f69a6804826e7bb07c43a797
+oid sha256:df1ea9b6a89f75f28de3af730e54e21e9f1d1b2dcd68114e100a9352120568a0
 size 5496