Initial alta tokenizer
Browse files- tokenizer.json +0 -0
- tokenizer_config.json +29 -0
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vocab_size": 65536,
|
| 3 |
+
"model_type": "BPE",
|
| 4 |
+
"byte_fallback": true,
|
| 5 |
+
"special_tokens": [
|
| 6 |
+
"<|pad|>",
|
| 7 |
+
"<|unk|>",
|
| 8 |
+
"<|bos|>",
|
| 9 |
+
"<|eos|>",
|
| 10 |
+
"<|im_start|>",
|
| 11 |
+
"<|im_end|>",
|
| 12 |
+
"<|begin_of_text|>",
|
| 13 |
+
"<|start_header_id|>",
|
| 14 |
+
"<|end_header_id|>",
|
| 15 |
+
"<|eot_id|>",
|
| 16 |
+
"<|thought_start|>",
|
| 17 |
+
"<|thought_end|>",
|
| 18 |
+
"<|tool_start|>",
|
| 19 |
+
"<|tool_end|>",
|
| 20 |
+
"<|tool_call|>",
|
| 21 |
+
"<|tool_result|>",
|
| 22 |
+
"<|fim_prefix|>",
|
| 23 |
+
"<|fim_middle|>",
|
| 24 |
+
"<|fim_suffix|>",
|
| 25 |
+
"<|fim_pad|>"
|
| 26 |
+
],
|
| 27 |
+
"language": "Kinyarwanda",
|
| 28 |
+
"apostrophe_handling": "preserved_in_words"
|
| 29 |
+
}
|