make vocabulary minuscule

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lgrobol/mbart-minuscule",
+  "_name_or_path": ".",
   "_num_labels": 3,
   "activation_dropout": 0.0,
   "activation_function": "relu",
@@ -45,7 +45,7 @@
   "normalize_before": true,
   "normalize_embedding": true,
   "num_beams": 5,
-  "num_hidden_layers": 12,
+  "num_hidden_layers": 2,
   "output_past": true,
   "pad_token_id": 1,
   "scale_embedding": true,
@@ -54,5 +54,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.26.1",
   "use_cache": true,
-  "vocab_size": 250054
+  "vocab_size": 256
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e21faf1a546468a8ff280281929734dde7244435145c040228bd5edb28597db
-size 131977105
+oid sha256:cb8abee1dcce4f5d4bed804f0447677245670384b90edcc733ba7a952be54bda
+size 3081361

special_tokens_map.json

@@ -53,10 +53,15 @@
     "gl_ES",
     "sl_SI"
   ],
-  "bos_token": "<s>",
   "cls_token": "<s>",
   "eos_token": "</s>",
-  "mask_token": "<mask>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
   "pad_token": "<pad>",
   "sep_token": "</s>",
   "unk_token": "<unk>"

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d58a68c276b56fcc48c165c63f70e5e4d452b4182032a5f7a2d018f4aa1a889
-size 17109752
+oid sha256:db4384eee58d54cf837101d35791567ebe1720e98e96b4e7c4a439af7ef4a0b7
+size 18600

tokenizer_config.json

@@ -53,10 +53,8 @@
     "gl_ES",
     "sl_SI"
   ],
-  "bos_token": "<s>",
   "cls_token": "<s>",
   "eos_token": "</s>",
-  "language_codes": "ML50",
   "mask_token": {
     "__type": "AddedToken",
     "content": "<mask>",
@@ -66,11 +64,8 @@
     "single_word": false
   },
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
   "pad_token": "<pad>",
   "sep_token": "</s>",
-  "sp_model_kwargs": {},
-  "special_tokens_map_file": "/home/suraj/projects/mbart-50/mbart-50/special_tokens_map.json",
   "src_lang": null,
   "tgt_lang": null,
   "tokenizer_class": "MBart50Tokenizer",