Upload 6 files

README.md

@@ -0,0 +1,28 @@
+---
+library_name: transformers
+tags:
+ - transformers.js
+ - tokenizers
+---
+
+# GPT-4 Tokenizer
+
+A 🤗-compatible version of the **GPT-4 tokenizer** (adapted from [openai/tiktoken](https://github.com/openai/tiktoken)). This means it can be used with Hugging Face libraries including [Transformers](https://github.com/huggingface/transformers), [Tokenizers](https://github.com/huggingface/tokenizers), and [Transformers.js](https://github.com/xenova/transformers.js).
+
+## Example usage:
+
+### Transformers/Tokenizers
+```py
+from transformers import GPT2TokenizerFast
+
+tokenizer = GPT2TokenizerFast.from_pretrained('Xenova/gpt-4')
+assert tokenizer.encode('hello world') == [15339, 1917]
+```
+
+### Transformers.js
+```js
+import { AutoTokenizer } from '@xenova/transformers';
+
+const tokenizer = await AutoTokenizer.from_pretrained('Xenova/gpt-4');
+const tokens = tokenizer.encode('hello world'); // [15339, 1917]
+```

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,10 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 8192,
+  "tokenizer_class": "GPT4Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}"
+}