Commit
•
4ada13f
1
Parent(s):
2c9dd72
Upload tokenizer
Browse files- added_tokens.json +6 -0
- special_tokens_map.json +6 -0
- spiece.model +3 -0
- tokenizer_config.json +15 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<cls>": 250102,
|
| 3 |
+
"<ctx>": 250100,
|
| 4 |
+
"<rot>": 250103,
|
| 5 |
+
"<sep>": 250101
|
| 6 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"sep_token": "<sep>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
|
| 3 |
+
size 4309802
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": null,
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"extra_ids": 0,
|
| 5 |
+
"model_max_length": 512,
|
| 6 |
+
"name_or_path": "google/mt5-base",
|
| 7 |
+
"pad_token": "<pad>",
|
| 8 |
+
"padding_side": "right",
|
| 9 |
+
"sp_model_kwargs": {},
|
| 10 |
+
"special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276",
|
| 11 |
+
"tokenizer_class": "T5Tokenizer",
|
| 12 |
+
"tokenizer_file": null,
|
| 13 |
+
"truncation_side": "right",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|