Training in progress, epoch 1
Browse files- added_tokens.json +4 -1
- config.json +1 -1
- pytorch_model.bin +2 -2
- special_tokens_map.json +5 -0
- tokenizer.json +27 -0
- tokenizer_config.json +29 -1
- training_args.bin +1 -1
added_tokens.json
CHANGED
|
@@ -3,5 +3,8 @@
|
|
| 3 |
"<mask>": 50264,
|
| 4 |
"<pad>": 1,
|
| 5 |
"<s>": 0,
|
| 6 |
-
"<unk>": 3
|
|
|
|
|
|
|
|
|
|
| 7 |
}
|
|
|
|
| 3 |
"<mask>": 50264,
|
| 4 |
"<pad>": 1,
|
| 5 |
"<s>": 0,
|
| 6 |
+
"<unk>": 3,
|
| 7 |
+
"[POSTTEXT]": 50266,
|
| 8 |
+
"[PRETEXT]": 50265,
|
| 9 |
+
"[TEXT]": 50267
|
| 10 |
}
|
config.json
CHANGED
|
@@ -32,5 +32,5 @@
|
|
| 32 |
"transformers_version": "4.34.0",
|
| 33 |
"type_vocab_size": 1,
|
| 34 |
"use_cache": true,
|
| 35 |
-
"vocab_size":
|
| 36 |
}
|
|
|
|
| 32 |
"transformers_version": "4.34.0",
|
| 33 |
"type_vocab_size": 1,
|
| 34 |
"use_cache": true,
|
| 35 |
+
"vocab_size": 50268
|
| 36 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3847de70b1b5a53defe4e8746f00d1b9a63c2627ce17a4a2daba722ee5079164
|
| 3 |
+
size 1421595057
|
special_tokens_map.json
CHANGED
|
@@ -1,4 +1,9 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"bos_token": "<s>",
|
| 3 |
"cls_token": "<s>",
|
| 4 |
"eos_token": "</s>",
|
|
|
|
| 1 |
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"[PRETEXT]",
|
| 4 |
+
"[POSTTEXT]",
|
| 5 |
+
"[TEXT]"
|
| 6 |
+
],
|
| 7 |
"bos_token": "<s>",
|
| 8 |
"cls_token": "<s>",
|
| 9 |
"eos_token": "</s>",
|
tokenizer.json
CHANGED
|
@@ -47,6 +47,33 @@
|
|
| 47 |
"rstrip": false,
|
| 48 |
"normalized": false,
|
| 49 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
}
|
| 51 |
],
|
| 52 |
"normalizer": null,
|
|
|
|
| 47 |
"rstrip": false,
|
| 48 |
"normalized": false,
|
| 49 |
"special": true
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"id": 50265,
|
| 53 |
+
"content": "[PRETEXT]",
|
| 54 |
+
"single_word": false,
|
| 55 |
+
"lstrip": true,
|
| 56 |
+
"rstrip": true,
|
| 57 |
+
"normalized": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"id": 50266,
|
| 62 |
+
"content": "[POSTTEXT]",
|
| 63 |
+
"single_word": false,
|
| 64 |
+
"lstrip": true,
|
| 65 |
+
"rstrip": true,
|
| 66 |
+
"normalized": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"id": 50267,
|
| 71 |
+
"content": "[TEXT]",
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"lstrip": true,
|
| 74 |
+
"rstrip": true,
|
| 75 |
+
"normalized": false,
|
| 76 |
+
"special": true
|
| 77 |
}
|
| 78 |
],
|
| 79 |
"normalizer": null,
|
tokenizer_config.json
CHANGED
|
@@ -40,9 +40,37 @@
|
|
| 40 |
"rstrip": false,
|
| 41 |
"single_word": false,
|
| 42 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
}
|
| 44 |
},
|
| 45 |
-
"additional_special_tokens": [
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
"bos_token": "<s>",
|
| 47 |
"clean_up_tokenization_spaces": true,
|
| 48 |
"cls_token": "<s>",
|
|
|
|
| 40 |
"rstrip": false,
|
| 41 |
"single_word": false,
|
| 42 |
"special": true
|
| 43 |
+
},
|
| 44 |
+
"50265": {
|
| 45 |
+
"content": "[PRETEXT]",
|
| 46 |
+
"lstrip": true,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": true,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"50266": {
|
| 53 |
+
"content": "[POSTTEXT]",
|
| 54 |
+
"lstrip": true,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": true,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"50267": {
|
| 61 |
+
"content": "[TEXT]",
|
| 62 |
+
"lstrip": true,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": true,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
}
|
| 68 |
},
|
| 69 |
+
"additional_special_tokens": [
|
| 70 |
+
"[PRETEXT]",
|
| 71 |
+
"[POSTTEXT]",
|
| 72 |
+
"[TEXT]"
|
| 73 |
+
],
|
| 74 |
"bos_token": "<s>",
|
| 75 |
"clean_up_tokenization_spaces": true,
|
| 76 |
"cls_token": "<s>",
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4091
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf95afb55fa2821768456fa3e30db1a2e0d6d04a968e1ffe8805cb92070e93ce
|
| 3 |
size 4091
|