Add model that's trained with zero-width characters

config.json

@@ -46,12 +46,12 @@
   "final_dropout": 0.0,
   "gradient_checkpointing": true,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.047,
+  "hidden_dropout": 0.05,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.041,
+  "layerdrop": 0.045,
   "mask_channel_length": 10,
   "mask_channel_min_space": 1,
   "mask_channel_other": 0.0,
@@ -70,7 +70,7 @@
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
   "num_hidden_layers": 24,
-  "pad_token_id": 73,
+  "pad_token_id": 75,
   "transformers_version": "4.5.0.dev0",
-  "vocab_size": 74
+  "vocab_size": 76
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e4efa4aa967aca324b06838550009a9d36cd249dfa6f8245a0d8ebe70c65777
-size 1262237207
+oid sha256:b5c39014b08099f35aa8eab263323ff24c61f959eddd9d50dc9716ab0df8afee
+size 1262245399

vocab.json

@@ -1 +1 @@
-{"ര": 0, "ജ": 1, "ഇ": 2, "ഝ": 3, "ആ": 4, "ൾ": 5, "ൌ": 6, "ത": 7, "ഠ": 8, "ണ": 9, "ദ": 10, "ധ": 11, "ശ": 12, "ഞ": 13, "ഐ": 14, "ി": 15, "വ": 16, "ൽ": 17, "ൻ": 18, "റ": 19, "യ": 20, "ഊ": 21, "ഛ": 22, "ു": 23, "ൃ": 24, "ഭ": 25, "ന": 26, "ർ": 27, "അ": 28, "ഘ": 29, "സ": 30, "ഗ": 31, "ൂ": 32, "ഃ": 33, "ഓ": 34, "ള": 35, "ല": 36, "ഋ": 37, "ഴ": 38, "ൗ": 39, "ം": 40, "ോ": 41, "ഥ": 42, "ഡ": 43, "ബ": 44, "െ": 45, "ാ": 47, "ൈ": 48, "ീ": 49, "ങ": 50, "ൊ": 51, "ഒ": 52, "േ": 53, "ക": 54, "ഷ": 55, "ഏ": 56, "ച": 57, "ഢ": 58, "ട": 59, "ഈ": 60, "എ": 61, "മ": 62, "്": 63, "ൺ": 64, "ഫ": 65, "ഖ": 66, "ഹ": 67, "ഉ": 68, "ഔ": 69, "പ": 70, "൱": 71, "|": 46, "[UNK]": 72, "[PAD]": 73}
+{"ർ": 0, "സ": 1, "ം": 2, "ഋ": 3, "ഹ": 4, "ധ": 5, "ഢ": 6, "ഠ": 7, "ട": 8, "പ": 9, "യ": 10, "ീ": 11, "ക": 12, "ര": 14, "ഏ": 15, "ഉ": 16, "ൃ": 17, "ൺ": 18, "ണ": 19, "ി": 20, "‌": 21, "ൽ": 22, "ൊ": 23, "ല": 24, "ൗ": 25, "ഘ": 26, "ൌ": 27, "ദ": 28, "ബ": 29, "ഛ": 30, "ഝ": 31, "ഊ": 32, "ന": 33, "ോ": 34, "൱": 35, "ഇ": 36, "ഐ": 37, "ൈ": 38, "ഃ": 39, "ാ": 40, "ച": 41, "േ": 42, "ഥ": 43, "ഡ": 44, "എ": 45, "ഷ": 46, "ആ": 47, "അ": 48, "മ": 49, "ഒ": 50, "ശ": 51, "ഖ": 52, "ഫ": 53, "ു": 54, "ഗ": 55, "ള": 56, "ഓ": 57, "ൂ": 58, "വ": 59, "ഞ": 60, "റ": 61, "ഭ": 62, "ൻ": 63, "ൾ": 64, "ഈ": 65, "്": 66, "െ": 67, "ങ": 68, "ത": 69, "ജ": 70, "‍": 71, "ഔ": 72, "ഴ": 73, "|": 13, "[UNK]": 74, "[PAD]": 75}