Merge branch 'main' into integration/sentence_transformers

README.md

@@ -2685,6 +2685,7 @@ sentences = ['What is TSNE?', 'Who is Laurens van der Maaten?']
 
 tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
 model = AutoModel.from_pretrained('nomic-ai/nomic-embed-text-v1', trust_remote_code=True)
+model.eval()
 
 encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
 

config.json

@@ -12,7 +12,7 @@
   "bos_token_id": null,
   "causal": false,
   "dense_seq_output": true,
-  "embd_pdrop": 0.1,
+  "embd_pdrop": 0.0,
   "eos_token_id": null,
   "fused_bias_fc": true,
   "fused_dropout_add_ln": true,
@@ -32,7 +32,7 @@
   "prenorm": false,
   "qkv_proj_bias": false,
   "reorder_and_upcast_attn": false,
-  "resid_pdrop": 0.1,
+  "resid_pdrop": 0.0,
   "rotary_emb_base": 1000,
   "rotary_emb_fraction": 1.0,
   "rotary_emb_interleaved": false,
@@ -41,7 +41,7 @@
   "scale_attn_by_inverse_layer_idx": false,
   "scale_attn_weights": true,
   "summary_activation": null,
-  "summary_first_dropout": 0.1,
+  "summary_first_dropout": 0.0,
   "summary_proj_to_labels": true,
   "summary_type": "cls_index",
   "summary_use_proj": true,