feat: add tokenizer

checkpoint-100/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f359164aeb72e09a6e708245c17ec33aaa3c2dfad171b1e42ee797caeaa7c21
+oid sha256:51837395b6badd36da9082d1e44666e9d7da54d0fe5c7a2d8d8ce682c5440ad8
 size 496250232

checkpoint-100/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48418ac8508208f573794a51b2290c7792ea29b31abbcfed445b9adb11b8aa03
+oid sha256:1b4911d0ae63127e97ccfffb9f8019fcb586b49fb2d589d325702775c6da237f
 size 992619066

checkpoint-100/trainer_state.json

@@ -11,9 +11,9 @@
   "log_history": [
     {
       "epoch": 50.0,
-      "grad_norm": 0.0506942979991436,
+      "grad_norm": 0.05403667315840721,
       "learning_rate": 2.0000000000000002e-07,
-      "loss": 0.1174,
+      "loss": 0.1177,
       "step": 100
     }
   ],

checkpoint-100/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ef21e526638efdea6d4da93396dad16ed9e50a73b427c476cd4f8423be070c9
+oid sha256:607ec35fd35c85bf5225edeebe4e07c5f5689c52764b3c323f98c47a9ba4395b
 size 5304