update model with 1.0 safeguard

checkpoint-68000/trainer_state.json

@@ -18,9 +18,9 @@
       "eval_bleu": 35.7559,
       "eval_gen_len": 39.7057,
       "eval_loss": 0.10894892364740372,
-      "eval_runtime": 186.2305,
-      "eval_samples_per_second": 5.601,
-      "eval_steps_per_second": 0.177,
+      "eval_runtime": 204.0367,
+      "eval_samples_per_second": 5.112,
+      "eval_steps_per_second": 0.162,
       "step": 4000
     },
     {
@@ -34,9 +34,9 @@
       "eval_bleu": 36.6369,
       "eval_gen_len": 39.395,
       "eval_loss": 0.10665024816989899,
-      "eval_runtime": 150.2548,
-      "eval_samples_per_second": 6.942,
-      "eval_steps_per_second": 0.22,
+      "eval_runtime": 169.0161,
+      "eval_samples_per_second": 6.171,
+      "eval_steps_per_second": 0.195,
       "step": 8000
     },
     {
@@ -50,9 +50,9 @@
       "eval_bleu": 41.2508,
       "eval_gen_len": 35.1687,
       "eval_loss": 0.10527685284614563,
-      "eval_runtime": 140.4867,
-      "eval_samples_per_second": 7.424,
-      "eval_steps_per_second": 0.235,
+      "eval_runtime": 161.3856,
+      "eval_samples_per_second": 6.463,
+      "eval_steps_per_second": 0.204,
       "step": 12000
     },
     {
@@ -66,9 +66,9 @@
       "eval_bleu": 41.4292,
       "eval_gen_len": 35.1361,
       "eval_loss": 0.104288749396801,
-      "eval_runtime": 117.5095,
-      "eval_samples_per_second": 8.876,
-      "eval_steps_per_second": 0.281,
+      "eval_runtime": 136.8172,
+      "eval_samples_per_second": 7.623,
+      "eval_steps_per_second": 0.241,
       "step": 16000
     },
     {
@@ -82,9 +82,9 @@
       "eval_bleu": 41.7716,
       "eval_gen_len": 35.5177,
       "eval_loss": 0.10370208323001862,
-      "eval_runtime": 146.3536,
-      "eval_samples_per_second": 7.127,
-      "eval_steps_per_second": 0.225,
+      "eval_runtime": 158.9983,
+      "eval_samples_per_second": 6.56,
+      "eval_steps_per_second": 0.208,
       "step": 20000
     },
     {
@@ -98,9 +98,9 @@
       "eval_bleu": 40.6317,
       "eval_gen_len": 36.8993,
       "eval_loss": 0.10362833738327026,
-      "eval_runtime": 131.4065,
-      "eval_samples_per_second": 7.937,
-      "eval_steps_per_second": 0.251,
+      "eval_runtime": 143.5271,
+      "eval_samples_per_second": 7.267,
+      "eval_steps_per_second": 0.23,
       "step": 24000
     },
     {
@@ -114,9 +114,9 @@
       "eval_bleu": 41.8852,
       "eval_gen_len": 35.3174,
       "eval_loss": 0.10282401740550995,
-      "eval_runtime": 124.1856,
-      "eval_samples_per_second": 8.399,
-      "eval_steps_per_second": 0.266,
+      "eval_runtime": 137.168,
+      "eval_samples_per_second": 7.604,
+      "eval_steps_per_second": 0.241,
       "step": 28000
     },
     {
@@ -130,9 +130,9 @@
       "eval_bleu": 41.5392,
       "eval_gen_len": 35.9501,
       "eval_loss": 0.10223321616649628,
-      "eval_runtime": 120.3748,
-      "eval_samples_per_second": 8.665,
-      "eval_steps_per_second": 0.274,
+      "eval_runtime": 133.1868,
+      "eval_samples_per_second": 7.831,
+      "eval_steps_per_second": 0.248,
       "step": 32000
     },
     {
@@ -146,9 +146,9 @@
       "eval_bleu": 40.5681,
       "eval_gen_len": 37.5935,
       "eval_loss": 0.10191329568624496,
-      "eval_runtime": 110.8617,
-      "eval_samples_per_second": 9.408,
-      "eval_steps_per_second": 0.298,
+      "eval_runtime": 125.5047,
+      "eval_samples_per_second": 8.31,
+      "eval_steps_per_second": 0.263,
       "step": 36000
     },
     {
@@ -162,9 +162,9 @@
       "eval_bleu": 42.3964,
       "eval_gen_len": 34.768,
       "eval_loss": 0.10204007476568222,
-      "eval_runtime": 112.6476,
-      "eval_samples_per_second": 9.259,
-      "eval_steps_per_second": 0.293,
+      "eval_runtime": 124.2655,
+      "eval_samples_per_second": 8.393,
+      "eval_steps_per_second": 0.266,
       "step": 40000
     },
     {
@@ -178,9 +178,9 @@
       "eval_bleu": 40.217,
       "eval_gen_len": 37.1946,
       "eval_loss": 0.10240339487791061,
-      "eval_runtime": 170.37,
-      "eval_samples_per_second": 6.122,
-      "eval_steps_per_second": 0.194,
+      "eval_runtime": 180.618,
+      "eval_samples_per_second": 5.775,
+      "eval_steps_per_second": 0.183,
       "step": 44000
     },
     {
@@ -194,9 +194,9 @@
       "eval_bleu": 41.5097,
       "eval_gen_len": 36.3797,
       "eval_loss": 0.1021459624171257,
-      "eval_runtime": 146.1468,
-      "eval_samples_per_second": 7.137,
-      "eval_steps_per_second": 0.226,
+      "eval_runtime": 149.0335,
+      "eval_samples_per_second": 6.998,
+      "eval_steps_per_second": 0.221,
       "step": 48000
     },
     {
@@ -210,9 +210,9 @@
       "eval_bleu": 42.2181,
       "eval_gen_len": 35.0019,
       "eval_loss": 0.10173720866441727,
-      "eval_runtime": 158.0705,
-      "eval_samples_per_second": 6.598,
-      "eval_steps_per_second": 0.209,
+      "eval_runtime": 168.836,
+      "eval_samples_per_second": 6.178,
+      "eval_steps_per_second": 0.195,
       "step": 52000
     },
     {
@@ -226,9 +226,9 @@
       "eval_bleu": 41.65,
       "eval_gen_len": 36.2694,
       "eval_loss": 0.10165542364120483,
-      "eval_runtime": 160.9591,
-      "eval_samples_per_second": 6.48,
-      "eval_steps_per_second": 0.205,
+      "eval_runtime": 181.2512,
+      "eval_samples_per_second": 5.754,
+      "eval_steps_per_second": 0.182,
       "step": 56000
     },
     {
@@ -242,9 +242,9 @@
       "eval_bleu": 42.5203,
       "eval_gen_len": 34.1112,
       "eval_loss": 0.10101941227912903,
-      "eval_runtime": 105.5932,
-      "eval_samples_per_second": 9.878,
-      "eval_steps_per_second": 0.313,
+      "eval_runtime": 117.7958,
+      "eval_samples_per_second": 8.854,
+      "eval_steps_per_second": 0.28,
       "step": 60000
     },
     {
@@ -258,9 +258,9 @@
       "eval_bleu": 42.2011,
       "eval_gen_len": 35.1908,
       "eval_loss": 0.10181207209825516,
-      "eval_runtime": 118.0171,
-      "eval_samples_per_second": 8.838,
-      "eval_steps_per_second": 0.28,
+      "eval_runtime": 136.2775,
+      "eval_samples_per_second": 7.654,
+      "eval_steps_per_second": 0.242,
       "step": 64000
     },
     {
@@ -274,9 +274,9 @@
       "eval_bleu": 42.5239,
       "eval_gen_len": 34.8715,
       "eval_loss": 0.10198543220758438,
-      "eval_runtime": 121.4795,
-      "eval_samples_per_second": 8.586,
-      "eval_steps_per_second": 0.272,
+      "eval_runtime": 134.6516,
+      "eval_samples_per_second": 7.746,
+      "eval_steps_per_second": 0.245,
       "step": 68000
     }
   ],

checkpoint-68000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb6b4d69457e8457bc8ba7157cadf6f7f462ddf893ce54473c85704df3bd5050
+oid sha256:4e6c61042e7a998c0fecc8637c8678e4c211809e57265d0edb7e047a99b13a43
 size 3771

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb6b4d69457e8457bc8ba7157cadf6f7f462ddf893ce54473c85704df3bd5050
+oid sha256:4e6c61042e7a998c0fecc8637c8678e4c211809e57265d0edb7e047a99b13a43
 size 3771