init tokenizer

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "unk_token": "<unk>",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f1255dbcb60721f8a1de89ab5e16e7d8702b8751ded3783c01108074f0a7d71
+size 2316877

tokenizer_config.json

@@ -0,0 +1,21 @@
+{
+  "tokenizer_class": "LlamaTokenizer",
+
+  "unk_token": "<unk>",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>",
+
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+
+  "padding_side": "left",
+
+  "clean_up_tokenization_spaces": false,
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "spaces_between_special_tokens": false,
+  "use_default_system_prompt": false
+}
+