add -v2 to README

README.md

@@ -8661,7 +8661,7 @@ Our new model that naturally integrates "context tokens" into the embedding proc
 <br>
 <hr>
 
-# How to use `cde-small-v1`
+# How to use `cde-small-v2`
 
 Our embedding model needs to be used in *two stages*. The first stage is to gather some dataset information by embedding a subset of the corpus using our "first-stage" model. The second stage is to actually embed queries and documents, conditioning on the corpus information from the first stage. Note that we can do the first stage part offline and only use the second-stage weights at inference time.
 
@@ -8670,7 +8670,7 @@ Our embedding model needs to be used in *two stages*. The first stage is to gath
 ## With Transformers
 
 <details>
-<summary>Click to learn how to use cde-small-v1 with Transformers</summary>
+<summary>Click to learn how to use cde-small-v2 with Transformers</summary>
 
 ### Loading the model
 
@@ -8678,7 +8678,7 @@ Our model can be loaded using `transformers` out-of-the-box with "trust remote c
 ```python
 import transformers
 
-model = transformers.AutoModel.from_pretrained("jxm/cde-small-v1", trust_remote_code=True)
+model = transformers.AutoModel.from_pretrained("jxm/cde-small-v2", trust_remote_code=True)
 tokenizer = transformers.AutoTokenizer.from_pretrained("bert-base-uncased")
 ```
 
@@ -8767,13 +8767,13 @@ these embeddings can be compared using dot product, since they're normalized.
 
 ### What if I don't know what my corpus will be ahead of time?
 
-If you can't obtain corpus information ahead of time, you still have to pass *something* as the dataset embeddings; our model will work fine in this case, but not quite as well; without corpus information, our model performance drops from 65.0 to 63.8 on MTEB. We provide [some random strings](https://huggingface.co/jxm/cde-small-v1/resolve/main/random_strings.txt) that worked well for us that can be used as a substitute for corpus sampling.
+If you can't obtain corpus information ahead of time, you still have to pass *something* as the dataset embeddings; our model will work fine in this case, but not quite as well; without corpus information, our model performance drops from 65.0 to 63.8 on MTEB. We provide [some random strings](https://huggingface.co/jxm/cde-small-v2/resolve/main/random_strings.txt) that worked well for us that can be used as a substitute for corpus sampling.
 
 
 ## With Sentence Transformers
 
 <details open="">
-<summary>Click to learn how to use cde-small-v1 with Sentence Transformers</summary>
+<summary>Click to learn how to use cde-small-v2 with Sentence Transformers</summary>
 
 ### Loading the model
 
@@ -8781,7 +8781,7 @@ Our model can be loaded using `sentence-transformers` out-of-the-box with "trust
 ```python
 from sentence_transformers import SentenceTransformer
 
-model = SentenceTransformer("jxm/cde-small-v1", trust_remote_code=True)
+model = SentenceTransformer("jxm/cde-small-v2", trust_remote_code=True)
 ```
 
 #### Note on prefixes
@@ -8838,7 +8838,7 @@ from sentence_transformers import SentenceTransformer
 from datasets import load_dataset
 
 # 1. Load the Sentence Transformer model
-model = SentenceTransformer("jxm/cde-small-v1", trust_remote_code=True)
+model = SentenceTransformer("jxm/cde-small-v2", trust_remote_code=True)
 context_docs_size = model[0].config.transductive_corpus_size  # 512
 
 # 2. Load the dataset: context dataset, docs, and queries