tokenizer stuff command line

added_tokens.json

@@ -0,0 +1,7 @@
+{
+  "<fim-middle>": 49154,
+  "<fim-pad>": 49156,
+  "<fim-prefix>": 49153,
+  "<fim-suffix>": 49155,
+  "<|endoftext|>": 49152
+}

special_tokens_map.json

@@ -0,0 +1,13 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim-prefix>",
+    "<fim-middle>",
+    "<fim-suffix>",
+    "<fim-pad>"
+  ],
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

tokenizer_config.json

@@ -0,0 +1,10 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 2048,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}