restage best simple adapt wce model

trainer_state.json

@@ -18,9 +18,9 @@
       "eval_bleu": 35.1893,
       "eval_gen_len": 40.3663,
       "eval_loss": 0.10897237807512283,
-      "eval_runtime": 192.3783,
-      "eval_samples_per_second": 5.422,
-      "eval_steps_per_second": 0.172,
+      "eval_runtime": 177.0634,
+      "eval_samples_per_second": 5.891,
+      "eval_steps_per_second": 0.186,
       "step": 4000
     },
     {
@@ -34,9 +34,9 @@
       "eval_bleu": 37.7679,
       "eval_gen_len": 37.7632,
       "eval_loss": 0.1064961776137352,
-      "eval_runtime": 177.2132,
-      "eval_samples_per_second": 5.886,
-      "eval_steps_per_second": 0.186,
+      "eval_runtime": 157.7862,
+      "eval_samples_per_second": 6.61,
+      "eval_steps_per_second": 0.209,
       "step": 8000
     },
     {
@@ -50,9 +50,9 @@
       "eval_bleu": 41.5566,
       "eval_gen_len": 34.2157,
       "eval_loss": 0.10526148974895477,
-      "eval_runtime": 149.8179,
-      "eval_samples_per_second": 6.962,
-      "eval_steps_per_second": 0.22,
+      "eval_runtime": 141.3674,
+      "eval_samples_per_second": 7.378,
+      "eval_steps_per_second": 0.233,
       "step": 12000
     },
     {
@@ -66,9 +66,9 @@
       "eval_bleu": 41.4171,
       "eval_gen_len": 35.5072,
       "eval_loss": 0.104159876704216,
-      "eval_runtime": 143.2125,
-      "eval_samples_per_second": 7.283,
-      "eval_steps_per_second": 0.23,
+      "eval_runtime": 123.7589,
+      "eval_samples_per_second": 8.428,
+      "eval_steps_per_second": 0.267,
       "step": 16000
     },
     {
@@ -82,9 +82,9 @@
       "eval_bleu": 41.7523,
       "eval_gen_len": 35.6031,
       "eval_loss": 0.10359229892492294,
-      "eval_runtime": 157.1246,
-      "eval_samples_per_second": 6.638,
-      "eval_steps_per_second": 0.21,
+      "eval_runtime": 148.6643,
+      "eval_samples_per_second": 7.016,
+      "eval_steps_per_second": 0.222,
       "step": 20000
     },
     {
@@ -98,9 +98,9 @@
       "eval_bleu": 41.5818,
       "eval_gen_len": 35.7987,
       "eval_loss": 0.10351784527301788,
-      "eval_runtime": 140.1257,
-      "eval_samples_per_second": 7.443,
-      "eval_steps_per_second": 0.236,
+      "eval_runtime": 131.6747,
+      "eval_samples_per_second": 7.921,
+      "eval_steps_per_second": 0.251,
       "step": 24000
     },
     {
@@ -114,9 +114,9 @@
       "eval_bleu": 41.7604,
       "eval_gen_len": 35.0968,
       "eval_loss": 0.1027546375989914,
-      "eval_runtime": 152.4696,
-      "eval_samples_per_second": 6.841,
-      "eval_steps_per_second": 0.216,
+      "eval_runtime": 137.6573,
+      "eval_samples_per_second": 7.577,
+      "eval_steps_per_second": 0.24,
       "step": 28000
     },
     {
@@ -130,9 +130,9 @@
       "eval_bleu": 41.9671,
       "eval_gen_len": 35.488,
       "eval_loss": 0.10217112302780151,
-      "eval_runtime": 134.38,
-      "eval_samples_per_second": 7.762,
-      "eval_steps_per_second": 0.246,
+      "eval_runtime": 124.3235,
+      "eval_samples_per_second": 8.389,
+      "eval_steps_per_second": 0.265,
       "step": 32000
     },
     {
@@ -146,9 +146,9 @@
       "eval_bleu": 41.2336,
       "eval_gen_len": 36.3039,
       "eval_loss": 0.1017456203699112,
-      "eval_runtime": 121.8994,
-      "eval_samples_per_second": 8.556,
-      "eval_steps_per_second": 0.271,
+      "eval_runtime": 112.1734,
+      "eval_samples_per_second": 9.298,
+      "eval_steps_per_second": 0.294,
       "step": 36000
     },
     {
@@ -162,9 +162,9 @@
       "eval_bleu": 42.312,
       "eval_gen_len": 34.4314,
       "eval_loss": 0.10187986493110657,
-      "eval_runtime": 125.3365,
-      "eval_samples_per_second": 8.322,
-      "eval_steps_per_second": 0.263,
+      "eval_runtime": 114.096,
+      "eval_samples_per_second": 9.141,
+      "eval_steps_per_second": 0.289,
       "step": 40000
     },
     {
@@ -178,9 +178,9 @@
       "eval_bleu": 42.1662,
       "eval_gen_len": 35.4765,
       "eval_loss": 0.10221881419420242,
-      "eval_runtime": 129.8395,
-      "eval_samples_per_second": 8.033,
-      "eval_steps_per_second": 0.254,
+      "eval_runtime": 123.1753,
+      "eval_samples_per_second": 8.468,
+      "eval_steps_per_second": 0.268,
       "step": 44000
     },
     {
@@ -194,9 +194,9 @@
       "eval_bleu": 42.1347,
       "eval_gen_len": 35.303,
       "eval_loss": 0.10202226787805557,
-      "eval_runtime": 130.0603,
-      "eval_samples_per_second": 8.019,
-      "eval_steps_per_second": 0.254,
+      "eval_runtime": 136.0863,
+      "eval_samples_per_second": 7.664,
+      "eval_steps_per_second": 0.242,
       "step": 48000
     },
     {
@@ -210,9 +210,9 @@
       "eval_bleu": 42.362,
       "eval_gen_len": 34.1179,
       "eval_loss": 0.10167563706636429,
-      "eval_runtime": 143.3476,
-      "eval_samples_per_second": 7.276,
-      "eval_steps_per_second": 0.23,
+      "eval_runtime": 138.7955,
+      "eval_samples_per_second": 7.515,
+      "eval_steps_per_second": 0.238,
       "step": 52000
     },
     {
@@ -226,9 +226,9 @@
       "eval_bleu": 42.0947,
       "eval_gen_len": 35.8955,
       "eval_loss": 0.10157355666160583,
-      "eval_runtime": 161.9595,
-      "eval_samples_per_second": 6.44,
-      "eval_steps_per_second": 0.204,
+      "eval_runtime": 162.7774,
+      "eval_samples_per_second": 6.408,
+      "eval_steps_per_second": 0.203,
       "step": 56000
     },
     {
@@ -242,9 +242,9 @@
       "eval_bleu": 42.504,
       "eval_gen_len": 34.5292,
       "eval_loss": 0.10091494768857956,
-      "eval_runtime": 144.7728,
-      "eval_samples_per_second": 7.204,
-      "eval_steps_per_second": 0.228,
+      "eval_runtime": 143.7894,
+      "eval_samples_per_second": 7.254,
+      "eval_steps_per_second": 0.23,
       "step": 60000
     },
     {
@@ -258,9 +258,9 @@
       "eval_bleu": 41.9688,
       "eval_gen_len": 33.8217,
       "eval_loss": 0.10167799890041351,
-      "eval_runtime": 122.082,
-      "eval_samples_per_second": 8.543,
-      "eval_steps_per_second": 0.27,
+      "eval_runtime": 128.1156,
+      "eval_samples_per_second": 8.141,
+      "eval_steps_per_second": 0.258,
       "step": 64000
     },
     {
@@ -274,9 +274,9 @@
       "eval_bleu": 42.6248,
       "eval_gen_len": 34.6213,
       "eval_loss": 0.1019180566072464,
-      "eval_runtime": 140.0619,
-      "eval_samples_per_second": 7.447,
-      "eval_steps_per_second": 0.236,
+      "eval_runtime": 136.0479,
+      "eval_samples_per_second": 7.666,
+      "eval_steps_per_second": 0.243,
       "step": 68000
     },
     {
@@ -290,9 +290,9 @@
       "eval_bleu": 42.3349,
       "eval_gen_len": 35.0364,
       "eval_loss": 0.10193591564893723,
-      "eval_runtime": 100.5506,
-      "eval_samples_per_second": 10.373,
-      "eval_steps_per_second": 0.328,
+      "eval_runtime": 101.1208,
+      "eval_samples_per_second": 10.314,
+      "eval_steps_per_second": 0.326,
       "step": 72000
     },
     {
@@ -306,9 +306,9 @@
       "eval_bleu": 41.4387,
       "eval_gen_len": 37.6692,
       "eval_loss": 0.10164961963891983,
-      "eval_runtime": 121.2832,
-      "eval_samples_per_second": 8.6,
-      "eval_steps_per_second": 0.272,
+      "eval_runtime": 125.4417,
+      "eval_samples_per_second": 8.315,
+      "eval_steps_per_second": 0.263,
       "step": 76000
     },
     {
@@ -322,9 +322,9 @@
       "eval_bleu": 42.9505,
       "eval_gen_len": 34.2924,
       "eval_loss": 0.10143885016441345,
-      "eval_runtime": 114.4227,
-      "eval_samples_per_second": 9.115,
-      "eval_steps_per_second": 0.288,
+      "eval_runtime": 116.3426,
+      "eval_samples_per_second": 8.965,
+      "eval_steps_per_second": 0.284,
       "step": 80000
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d092665eee12a3fa4a88455c50bbbac4daa679ca877ebc30400154070acc871
+oid sha256:d84e5a6c16cedb85a269964ac0820de47b0330360a6c003d0b08a8f090d7c49e
 size 3835