Training in progress, step 1100

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33f7cf441493e72909629a82d2c0fe91b53e0be257030ae6bd2508f535036fef
+oid sha256:f2eda2395cb081ee600d9dd06500d6a1fe8343ddcc1b00978b95e0d6b766d525
 size 496295078

trainer_state.json

@@ -153,22 +153,22 @@
       "step": 1000,
       "total_flos": 6.688961805360538e+16,
       "train_loss": 0.0,
-      "train_runtime": 0.614,
-      "train_samples_per_second": 212612.275,
-      "train_steps_per_second": 415.287
+      "train_runtime": 2.2787,
+      "train_samples_per_second": 229166.609,
+      "train_steps_per_second": 222.933
     },
     {
       "epoch": 1.96,
-      "eval_loss": 0.9063528180122375,
-      "eval_runtime": 17.3907,
-      "eval_samples_per_second": 687.437,
-      "eval_steps_per_second": 5.405,
+      "eval_loss": 0.9063556790351868,
+      "eval_runtime": 17.1279,
+      "eval_samples_per_second": 697.982,
+      "eval_steps_per_second": 2.744,
       "step": 1000
     }
   ],
   "logging_steps": 100,
-  "max_steps": 255,
-  "num_train_epochs": 1,
+  "max_steps": 508,
+  "num_train_epochs": 4,
   "save_steps": 100,
   "total_flos": 6.688961805360538e+16,
   "trial_name": null,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1776a22620d47e79406d32e398cf9107341a37c14af882ed10b1cef99f96d914
+oid sha256:5272da2cb97870f8ea42ccff3371d5029829e7363293d9a07c34074955cd23e8
 size 4664