Add Sentence Transformers support (#2)

- Add return_dict=None as required by Sentence Transformers (f91d416e5dbd84ced31f595672eac67c4b785972)
- Update tokenizer_config: max length 512 -> 8192 (6b2747813ff1e50931b8f9f3c70d4cc42ff4f0b6)
- Add Sentence Transformers files (af02f33ae73c8a1b6ba5b3119ae641b51e52cc74)
- Add Sentence Transformers usage snippet (8411d8efe1cef1bb1a167b47b76bf912a3b60e51)
- Add Sentence Transformers metadata (78b8337bcc2d997c837ec71324838636a538fd83)

1_Pooling/config.json

@@ -1,10 +1,9 @@
 {
-    "word_embedding_dimension": 768,
-    "pooling_mode_cls_token": false,
-    "pooling_mode_mean_tokens": true,
-    "pooling_mode_max_tokens": false,
-    "pooling_mode_mean_sqrt_len_tokens": false,
-    "pooling_mode_weightedmean_tokens": false,
-    "pooling_mode_lasttoken": false
-  }
-  
+  "word_embedding_dimension": 768,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false
+}

README.md

@@ -1,6 +1,10 @@
 ---
+library_name: sentence-transformers
+pipeline_tag: sentence-similarity
 tags:
-- mteb
+  - feature-extraction
+  - sentence-similarity
+  - mteb
 model-index:
 - name: epoch_0_model
   results:
@@ -2653,6 +2657,17 @@ Training data to train the models is released in its entirety. For more details,
 
 ## Usage
 
+### Sentence Transformers
+```python
+from sentence_transformers import SentenceTransformer
+
+model = SentenceTransformer("nomic-ai/nomic-embed-text-v1-ablated", trust_remote_code=True)
+sentences = ['What is TSNE?', 'Who is Laurens van der Maaten?']
+embeddings = model.encode(sentences)
+print(embeddings)
+```
+
+### Transformers
 
 ```python
 import torch

config_sentence_transformers.json

@@ -0,0 +1,7 @@
+{
+  "__version__": {
+    "sentence_transformers": "2.4.0.dev0",
+    "transformers": "4.37.2",
+    "pytorch": "2.1.0+cu121"
+  }
+}

modeling_hf_nomic_bert.py

@@ -1069,6 +1069,7 @@ class NomicBertModel(NomicBertPreTrainedModel):
         position_ids=None,
         token_type_ids=None,
         attention_mask=None,
+        return_dict=None,
     ):
         if token_type_ids is None:
             token_type_ids = torch.zeros_like(input_ids)
@@ -1080,7 +1081,7 @@ class NomicBertModel(NomicBertPreTrainedModel):
 
         attention_mask = self.get_extended_attention_mask(attention_mask, input_ids.shape)
         sequence_output = self.encoder(
-            hidden_states, attention_mask=attention_mask
+            hidden_states, attention_mask=attention_mask, return_dict=return_dict,
         )
 
         pooled_output = self.pooler(sequence_output) if self.pooler is not None else None

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 8192,
+  "do_lower_case": false
+}

tokenizer_config.json

@@ -46,7 +46,7 @@
   "cls_token": "[CLS]",
   "do_lower_case": true,
   "mask_token": "[MASK]",
-  "model_max_length": 512,
+  "model_max_length": 8192,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
   "strip_accents": null,