Model save

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 2.9978431433840766,
     "total_flos": 1.9278929080237425e+18,
-    "train_loss": 0.3391870107895778,
-    "train_runtime": 31542.9957,
+    "train_loss": 0.3418351112304626,
+    "train_runtime": 31635.6273,
     "train_samples": 63982,
-    "train_samples_per_second": 6.085,
+    "train_samples_per_second": 6.067,
     "train_steps_per_second": 0.095
 }

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a48d694a29804eb9d12b8ced279b542d707186f3daf6cff8c8c5d9200f14e2b4
+oid sha256:e94664ad14df07d014ff7c61583ac9ecf893ed67ff31b8657ed54fcdcfd667d9
 size 4976698672

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c34e245f8cd6490cc3b360786a826fa9fc1b394bcf65b2261dc6bf1c0ae754bb
+oid sha256:fbd7af2e261a5615659698f29ea28cc68f7d338f40cbec3bfe3daa558a1ed97c
 size 4999802720

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1bfca41224c5e7da9448e00e5bb4cb8cb21aee969e78f1af1e873f2a17dc499
+oid sha256:390b29ca3b5036d1e837b0029938c275b1c4d41bff8f3b200fb239d2f03db503
 size 4915916176

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32a2539719eb5119644cea1e70c1291b614162016924ff31584f03183f2a6cb2
+oid sha256:3eb9a2caf299496568bba1836c833f5455ac8f8079bf142a3912ac77eb9c2f75
 size 1168138808

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 2.9978431433840766,
     "total_flos": 1.9278929080237425e+18,
-    "train_loss": 0.3391870107895778,
-    "train_runtime": 31542.9957,
+    "train_loss": 0.3418351112304626,
+    "train_runtime": 31635.6273,
     "train_samples": 63982,
-    "train_samples_per_second": 6.085,
+    "train_samples_per_second": 6.067,
     "train_steps_per_second": 0.095
 }

trainer_state.json

@@ -17,424 +17,424 @@
     },
     {
       "epoch": 0.0500140664561908,
-      "grad_norm": 6.46875,
+      "grad_norm": 9.0625,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 1.2485,
+      "loss": 1.2593,
       "step": 50
     },
     {
       "epoch": 0.1000281329123816,
-      "grad_norm": 3.296875,
+      "grad_norm": 4.1875,
       "learning_rate": 6.666666666666667e-06,
-      "loss": 0.6295,
+      "loss": 0.6353,
       "step": 100
     },
     {
       "epoch": 0.1500421993685724,
-      "grad_norm": 3.40625,
+      "grad_norm": 3.4375,
       "learning_rate": 1e-05,
-      "loss": 0.567,
+      "loss": 0.5783,
       "step": 150
     },
     {
       "epoch": 0.2000562658247632,
-      "grad_norm": 3.890625,
+      "grad_norm": 3.09375,
       "learning_rate": 1.3333333333333333e-05,
-      "loss": 0.6028,
+      "loss": 0.5614,
       "step": 200
     },
     {
       "epoch": 0.25007033228095404,
-      "grad_norm": 2.84375,
+      "grad_norm": 3.421875,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.6082,
+      "loss": 0.6185,
       "step": 250
     },
     {
       "epoch": 0.3000843987371448,
-      "grad_norm": 2.96875,
+      "grad_norm": 7.5,
       "learning_rate": 2e-05,
-      "loss": 0.5956,
+      "loss": 0.6262,
       "step": 300
     },
     {
       "epoch": 0.35009846519333565,
-      "grad_norm": 2.203125,
+      "grad_norm": 2.421875,
       "learning_rate": 1.9983043934122208e-05,
-      "loss": 0.5838,
+      "loss": 0.5878,
       "step": 350
     },
     {
       "epoch": 0.4001125316495264,
-      "grad_norm": 2.5625,
+      "grad_norm": 3.171875,
       "learning_rate": 1.9932233238122834e-05,
-      "loss": 0.5634,
+      "loss": 0.575,
       "step": 400
     },
     {
       "epoch": 0.45012659810571726,
       "grad_norm": 2.0,
       "learning_rate": 1.984774022190361e-05,
-      "loss": 0.5472,
+      "loss": 0.5526,
       "step": 450
     },
     {
       "epoch": 0.5001406645619081,
-      "grad_norm": 1.84375,
+      "grad_norm": 1.8203125,
       "learning_rate": 1.972985141929439e-05,
-      "loss": 0.5359,
+      "loss": 0.54,
       "step": 500
     },
     {
       "epoch": 0.5501547310180989,
-      "grad_norm": 1.9375,
+      "grad_norm": 2.03125,
       "learning_rate": 1.9578966616355823e-05,
-      "loss": 0.5243,
+      "loss": 0.527,
       "step": 550
     },
     {
       "epoch": 0.6001687974742896,
-      "grad_norm": 1.9453125,
+      "grad_norm": 2.046875,
       "learning_rate": 1.9395597495619634e-05,
-      "loss": 0.5194,
+      "loss": 0.5229,
       "step": 600
     },
     {
       "epoch": 0.6501828639304804,
-      "grad_norm": 1.578125,
+      "grad_norm": 1.734375,
       "learning_rate": 1.918036590086405e-05,
-      "loss": 0.5046,
+      "loss": 0.5062,
       "step": 650
     },
     {
       "epoch": 0.7001969303866713,
-      "grad_norm": 1.7265625,
+      "grad_norm": 1.578125,
       "learning_rate": 1.8934001728309003e-05,
-      "loss": 0.5029,
+      "loss": 0.5055,
       "step": 700
     },
     {
       "epoch": 0.7502109968428621,
-      "grad_norm": 1.703125,
+      "grad_norm": 1.7265625,
       "learning_rate": 1.865734045138245e-05,
-      "loss": 0.4931,
+      "loss": 0.4947,
       "step": 750
     },
     {
       "epoch": 0.8002250632990529,
-      "grad_norm": 1.5625,
+      "grad_norm": 1.5390625,
       "learning_rate": 1.8351320287451865e-05,
-      "loss": 0.4897,
+      "loss": 0.491,
       "step": 800
     },
     {
       "epoch": 0.8502391297552436,
-      "grad_norm": 1.390625,
+      "grad_norm": 1.4765625,
       "learning_rate": 1.8016979016129164e-05,
-      "loss": 0.4807,
+      "loss": 0.4824,
       "step": 850
     },
     {
       "epoch": 0.9002531962114345,
-      "grad_norm": 1.515625,
+      "grad_norm": 1.53125,
       "learning_rate": 1.7655450459938786e-05,
-      "loss": 0.4724,
+      "loss": 0.4738,
       "step": 900
     },
     {
       "epoch": 0.9502672626676253,
-      "grad_norm": 1.5,
+      "grad_norm": 1.65625,
       "learning_rate": 1.726796063928382e-05,
-      "loss": 0.4654,
+      "loss": 0.4676,
       "step": 950
     },
     {
       "epoch": 1.0002813291238162,
-      "grad_norm": 1.4296875,
+      "grad_norm": 1.40625,
       "learning_rate": 1.6855823614749474e-05,
-      "loss": 0.4645,
+      "loss": 0.4657,
       "step": 1000
     },
     {
       "epoch": 1.050295395580007,
-      "grad_norm": 1.3359375,
+      "grad_norm": 1.40625,
       "learning_rate": 1.6420437030843482e-05,
-      "loss": 0.3193,
+      "loss": 0.3223,
       "step": 1050
     },
     {
       "epoch": 1.1003094620361977,
-      "grad_norm": 1.3125,
+      "grad_norm": 1.3359375,
       "learning_rate": 1.5963277376285646e-05,
-      "loss": 0.3168,
+      "loss": 0.3197,
       "step": 1100
     },
     {
       "epoch": 1.1503235284923885,
-      "grad_norm": 1.3046875,
+      "grad_norm": 1.2890625,
       "learning_rate": 1.5485894976919836e-05,
-      "loss": 0.3189,
+      "loss": 0.3246,
       "step": 1150
     },
     {
       "epoch": 1.2003375949485793,
-      "grad_norm": 1.3984375,
+      "grad_norm": 1.421875,
       "learning_rate": 1.4989908738228567e-05,
-      "loss": 0.3143,
+      "loss": 0.3167,
       "step": 1200
     },
     {
       "epoch": 1.25035166140477,
-      "grad_norm": 1.3515625,
+      "grad_norm": 1.3828125,
       "learning_rate": 1.4477000655279376e-05,
-      "loss": 0.3163,
+      "loss": 0.3186,
       "step": 1250
     },
     {
       "epoch": 1.3003657278609608,
-      "grad_norm": 1.34375,
+      "grad_norm": 1.3515625,
       "learning_rate": 1.394891010872102e-05,
-      "loss": 0.3129,
+      "loss": 0.3149,
       "step": 1300
     },
     {
       "epoch": 1.3503797943171518,
-      "grad_norm": 1.40625,
+      "grad_norm": 1.3828125,
       "learning_rate": 1.3407427966172866e-05,
-      "loss": 0.3133,
+      "loss": 0.3164,
       "step": 1350
     },
     {
       "epoch": 1.4003938607733426,
-      "grad_norm": 1.3515625,
+      "grad_norm": 1.34375,
       "learning_rate": 1.2854390509011061e-05,
-      "loss": 0.3103,
+      "loss": 0.313,
       "step": 1400
     },
     {
       "epoch": 1.4504079272295334,
-      "grad_norm": 1.3203125,
+      "grad_norm": 1.390625,
       "learning_rate": 1.2291673205146908e-05,
-      "loss": 0.3051,
+      "loss": 0.3071,
       "step": 1450
     },
     {
       "epoch": 1.5004219936857242,
-      "grad_norm": 1.3046875,
+      "grad_norm": 1.328125,
       "learning_rate": 1.1721184348915384e-05,
-      "loss": 0.3043,
+      "loss": 0.3063,
       "step": 1500
     },
     {
       "epoch": 1.550436060141915,
-      "grad_norm": 1.2734375,
+      "grad_norm": 1.2890625,
       "learning_rate": 1.1144858589642251e-05,
-      "loss": 0.2998,
+      "loss": 0.3022,
       "step": 1550
     },
     {
       "epoch": 1.6004501265981057,
-      "grad_norm": 1.3046875,
+      "grad_norm": 1.3359375,
       "learning_rate": 1.0564650370835772e-05,
-      "loss": 0.2983,
+      "loss": 0.3006,
       "step": 1600
     },
     {
       "epoch": 1.6504641930542965,
-      "grad_norm": 1.34375,
+      "grad_norm": 1.3359375,
       "learning_rate": 9.982527302252135e-06,
-      "loss": 0.297,
+      "loss": 0.2998,
       "step": 1650
     },
     {
       "epoch": 1.7004782595104873,
       "grad_norm": 1.359375,
       "learning_rate": 9.40046348731131e-06,
-      "loss": 0.2918,
+      "loss": 0.2947,
       "step": 1700
     },
     {
       "epoch": 1.750492325966678,
-      "grad_norm": 1.40625,
+      "grad_norm": 1.34375,
       "learning_rate": 8.820432828491542e-06,
-      "loss": 0.2919,
+      "loss": 0.294,
       "step": 1750
     },
     {
       "epoch": 1.8005063924228688,
-      "grad_norm": 1.34375,
+      "grad_norm": 1.375,
       "learning_rate": 8.244402333405252e-06,
-      "loss": 0.2872,
+      "loss": 0.2894,
       "step": 1800
     },
     {
       "epoch": 1.8505204588790596,
-      "grad_norm": 1.328125,
+      "grad_norm": 1.375,
       "learning_rate": 7.674325444256899e-06,
-      "loss": 0.2857,
+      "loss": 0.2879,
       "step": 1850
     },
     {
       "epoch": 1.9005345253352506,
-      "grad_norm": 1.390625,
+      "grad_norm": 1.3671875,
       "learning_rate": 7.112135413304042e-06,
-      "loss": 0.2821,
+      "loss": 0.2842,
       "step": 1900
     },
     {
       "epoch": 1.9505485917914414,
-      "grad_norm": 1.2421875,
+      "grad_norm": 1.3125,
       "learning_rate": 6.55973874678682e-06,
-      "loss": 0.2812,
+      "loss": 0.2839,
       "step": 1950
     },
     {
       "epoch": 2.0005626582476324,
-      "grad_norm": 1.09375,
+      "grad_norm": 1.1171875,
       "learning_rate": 6.0190087395588596e-06,
-      "loss": 0.2739,
+      "loss": 0.2766,
       "step": 2000
     },
     {
       "epoch": 2.050576724703823,
-      "grad_norm": 1.2265625,
+      "grad_norm": 1.3046875,
       "learning_rate": 5.491779122345093e-06,
-      "loss": 0.1498,
+      "loss": 0.1517,
       "step": 2050
     },
     {
       "epoch": 2.100590791160014,
-      "grad_norm": 1.1796875,
+      "grad_norm": 1.1953125,
       "learning_rate": 4.979837843169959e-06,
-      "loss": 0.1477,
+      "loss": 0.1492,
       "step": 2100
     },
     {
       "epoch": 2.1506048576162047,
-      "grad_norm": 1.1953125,
+      "grad_norm": 1.203125,
       "learning_rate": 4.484921004044509e-06,
-      "loss": 0.1482,
+      "loss": 0.1494,
       "step": 2150
     },
     {
       "epoch": 2.2006189240723955,
-      "grad_norm": 1.140625,
+      "grad_norm": 1.1484375,
       "learning_rate": 4.008706973474391e-06,
-      "loss": 0.1483,
+      "loss": 0.1498,
       "step": 2200
     },
     {
       "epoch": 2.2506329905285862,
       "grad_norm": 1.1328125,
       "learning_rate": 3.5528106947544626e-06,
-      "loss": 0.1464,
+      "loss": 0.1482,
       "step": 2250
     },
     {
       "epoch": 2.300647056984777,
-      "grad_norm": 1.1953125,
+      "grad_norm": 1.1796875,
       "learning_rate": 3.118778209351808e-06,
-      "loss": 0.1463,
+      "loss": 0.1477,
       "step": 2300
     },
     {
       "epoch": 2.350661123440968,
-      "grad_norm": 1.3125,
+      "grad_norm": 1.234375,
       "learning_rate": 2.7080814139495402e-06,
-      "loss": 0.1456,
+      "loss": 0.1473,
       "step": 2350
     },
     {
       "epoch": 2.4006751898971586,
-      "grad_norm": 1.3046875,
+      "grad_norm": 1.2421875,
       "learning_rate": 2.322113068931391e-06,
-      "loss": 0.1453,
+      "loss": 0.147,
       "step": 2400
     },
     {
       "epoch": 2.4506892563533493,
       "grad_norm": 1.1875,
       "learning_rate": 1.9621820752343324e-06,
-      "loss": 0.145,
+      "loss": 0.1466,
       "step": 2450
     },
     {
       "epoch": 2.50070332280954,
-      "grad_norm": 1.203125,
+      "grad_norm": 1.21875,
       "learning_rate": 1.629509035586484e-06,
-      "loss": 0.1436,
+      "loss": 0.145,
       "step": 2500
     },
     {
       "epoch": 2.550717389265731,
-      "grad_norm": 1.296875,
+      "grad_norm": 1.4140625,
       "learning_rate": 1.3252221151830513e-06,
-      "loss": 0.1439,
+      "loss": 0.146,
       "step": 2550
     },
     {
       "epoch": 2.6007314557219217,
-      "grad_norm": 1.203125,
+      "grad_norm": 1.15625,
       "learning_rate": 1.0503532158376584e-06,
-      "loss": 0.1432,
+      "loss": 0.1453,
       "step": 2600
     },
     {
       "epoch": 2.6507455221781124,
       "grad_norm": 1.203125,
       "learning_rate": 8.058344765833171e-07,
-      "loss": 0.1442,
+      "loss": 0.1466,
       "step": 2650
     },
     {
       "epoch": 2.7007595886343037,
-      "grad_norm": 1.2109375,
+      "grad_norm": 1.2734375,
       "learning_rate": 5.924951125902545e-07,
-      "loss": 0.1443,
+      "loss": 0.1458,
       "step": 2700
     },
     {
       "epoch": 2.7507736550904944,
-      "grad_norm": 1.171875,
+      "grad_norm": 1.1640625,
       "learning_rate": 4.11058603120511e-07,
-      "loss": 0.1428,
+      "loss": 0.1445,
       "step": 2750
     },
     {
       "epoch": 2.800787721546685,
-      "grad_norm": 1.1640625,
+      "grad_norm": 1.1875,
       "learning_rate": 2.6214023805552826e-07,
-      "loss": 0.1432,
+      "loss": 0.1449,
       "step": 2800
     },
     {
       "epoch": 2.850801788002876,
-      "grad_norm": 1.2578125,
+      "grad_norm": 1.390625,
       "learning_rate": 1.462450313169983e-07,
-      "loss": 0.1434,
+      "loss": 0.1457,
       "step": 2850
     },
     {
       "epoch": 2.9008158544590668,
-      "grad_norm": 1.1484375,
+      "grad_norm": 1.2265625,
       "learning_rate": 6.376600825699463e-08,
-      "loss": 0.1427,
+      "loss": 0.1444,
       "step": 2900
     },
     {
       "epoch": 2.9508299209152575,
-      "grad_norm": 1.1796875,
+      "grad_norm": 1.171875,
       "learning_rate": 1.49828728252277e-08,
-      "loss": 0.1438,
+      "loss": 0.1451,
       "step": 2950
     },
     {
       "epoch": 2.9978431433840766,
       "step": 2997,
       "total_flos": 1.9278929080237425e+18,
-      "train_loss": 0.3391870107895778,
-      "train_runtime": 31542.9957,
-      "train_samples_per_second": 6.085,
+      "train_loss": 0.3418351112304626,
+      "train_runtime": 31635.6273,
+      "train_samples_per_second": 6.067,
       "train_steps_per_second": 0.095
     }
   ],