codebyzeb commited on
Commit
88b7b77
·
verified ·
1 Parent(s): 7f9bd40

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -62,3 +62,4 @@ frequencymultirussian_128000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
62
  frequencymultirussian_256000/merges_data.csv filter=lfs diff=lfs merge=lfs -text
63
  frequencymultirussian_256000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
64
  frequencymultirussian_256000/vocab.json filter=lfs diff=lfs merge=lfs -text
 
 
62
  frequencymultirussian_256000/merges_data.csv filter=lfs diff=lfs merge=lfs -text
63
  frequencymultirussian_256000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
64
  frequencymultirussian_256000/vocab.json filter=lfs diff=lfs merge=lfs -text
65
+ fw57M_Surprisal_thresholdB_64000/stats.csv filter=lfs diff=lfs merge=lfs -text
fw57M_Surprisal_thresholdB_64000/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|padding|>",
5
+ "unk_token": "<|unk|>"
6
+ }
fw57M_Surprisal_thresholdB_64000/stats.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd47d4646776cfe19384f4582184a655a2c057c432f084edbf1726cd048d5fc8
3
+ size 14255327
fw57M_Surprisal_thresholdB_64000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fw57M_Surprisal_thresholdB_64000/tokenizer_config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<|padding|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<|endoftext|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "698": {
21
+ "content": "<|unk|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ }
28
+ },
29
+ "bos_token": "<|endoftext|>",
30
+ "clean_up_tokenization_spaces": false,
31
+ "eos_token": "<|endoftext|>",
32
+ "extra_special_tokens": {},
33
+ "model_max_length": 1000000000000000019884624838656,
34
+ "pad_token": "<|padding|>",
35
+ "tokenizer_class": "PreTrainedTokenizer",
36
+ "unk_token": "<|unk|>"
37
+ }
fw57M_Surprisal_thresholdB_64000/vocab.json ADDED
The diff for this file is too large to render. See raw diff