Hugging Face's logo

divakantpandey
/
code-embed

Feature Extraction
sentence-transformers
ONNX
Safetensors
English
code
roberta
python
code-search
code-embeddings
unixcoder
matryoshka
mteb
Eval Results (legacy)
text-embeddings-inference
Model card Files
xet
 Community
divakantpandey commited on
Commit
0f0286c
·
verified ·
1 Parent(s): 13aff62

fix: strictly cap max_seq_length and model_max_length to 8192 for RoBERTa position embeddings

Browse files
Files changed (8) hide show
  1. backbone/tokenizer.json +0 -0
  2. backbone/tokenizer_config.json +24 -0
  3. code_embed_config.json +1 -1
  4. projection_head.pt +1 -1
  5. pycode_embed_config.json +1 -1
  6. sentence_bert_config.json +9 -8
  7. tokenizer.json +6 -1
  8. tokenizer_config.json +2 -2
backbone/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
backbone/tokenizer_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "is_local": true,
9
+ "local_files_only": false,
10
+ "mask_token": "<mask>",
11
+ "max_length": 8192,
12
+ "model_max_length": 8192,
13
+ "pad_to_multiple_of": null,
14
+ "pad_token": "<pad>",
15
+ "pad_token_type_id": 0,
16
+ "padding_side": "right",
17
+ "sep_token": "</s>",
18
+ "stride": 0,
19
+ "tokenizer_class": "RobertaTokenizer",
20
+ "trim_offsets": true,
21
+ "truncation_side": "right",
22
+ "truncation_strategy": "longest_first",
23
+ "unk_token": "<unk>"
24
+ }
code_embed_config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "base_model_name_or_path": "microsoft/unixcoder-base",
3
  "backbone_hidden_size": 768,
4
  "embedding_dim": 1536,
5
  "projection_hidden_dim": 1536,
 
1
  {
2
+ "base_model_name_or_path": "outputs/best_model/backbone",
3
  "backbone_hidden_size": 768,
4
  "embedding_dim": 1536,
5
  "projection_hidden_dim": 1536,
projection_head.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cbf1e13bb707299ddc99b72622c9807e48d44a58c72cbd399e61e081abfca2e4
3
  size 14170669
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:617dd6118c979066054e8ac303613370c9553c869c440d3c7336416854325c6d
3
  size 14170669
pycode_embed_config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "base_model_name_or_path": "microsoft/unixcoder-base",
3
  "backbone_hidden_size": 768,
4
  "embedding_dim": 1536,
5
  "projection_hidden_dim": 1536,
 
1
  {
2
+ "base_model_name_or_path": "outputs/best_model/backbone",
3
  "backbone_hidden_size": 768,
4
  "embedding_dim": 1536,
5
  "projection_hidden_dim": 1536,
sentence_bert_config.json CHANGED
@@ -1,10 +1,11 @@
1
  {
2
- "transformer_task": "feature-extraction",
3
- "modality_config": {
4
- "text": {
5
- "method": "forward",
6
- "method_output_name": "last_hidden_state"
7
- }
8
- },
9
- "module_output_name": "token_embeddings"
 
10
  }
 
1
  {
2
+ "transformer_task": "feature-extraction",
3
+ "modality_config": {
4
+ "text": {
5
+ "method": "forward",
6
+ "method_output_name": "last_hidden_state"
7
+ }
8
+ },
9
+ "module_output_name": "token_embeddings",
10
+ "max_seq_length": 8192
11
  }
tokenizer.json CHANGED
@@ -1,6 +1,11 @@
1
  {
2
  "version": "1.0",
3
- "truncation": null,
 
 
 
 
 
4
  "padding": {
5
  "strategy": "BatchLongest",
6
  "direction": "Right",
 
1
  {
2
  "version": "1.0",
3
+ "truncation": {
4
+ "direction": "Right",
5
+ "max_length": 8192,
6
+ "strategy": "LongestFirst",
7
+ "stride": 0
8
+ },
9
  "padding": {
10
  "strategy": "BatchLongest",
11
  "direction": "Right",
tokenizer_config.json CHANGED
@@ -9,7 +9,7 @@
9
  "local_files_only": false,
10
  "mask_token": "<mask>",
11
  "max_length": 8192,
12
- "model_max_length": 1000000000000000019884624838656,
13
  "pad_to_multiple_of": null,
14
  "pad_token": "<pad>",
15
  "pad_token_type_id": 0,
@@ -21,4 +21,4 @@
21
  "truncation_side": "right",
22
  "truncation_strategy": "longest_first",
23
  "unk_token": "<unk>"
24
- }
 
9
  "local_files_only": false,
10
  "mask_token": "<mask>",
11
  "max_length": 8192,
12
+ "model_max_length": 8192,
13
  "pad_to_multiple_of": null,
14
  "pad_token": "<pad>",
15
  "pad_token_type_id": 0,
 
21
  "truncation_side": "right",
22
  "truncation_strategy": "longest_first",
23
  "unk_token": "<unk>"
24
+ }