Update text_encoder/config.json

text_encoder/config.json

@@ -1,4 +1,5 @@
 {
+  "_name_or_path": "laion/CLIP-ViT-H-14-laion2B-s32B-b79K",
   "architectures": [
     "CLIPTextModel"
   ],
@@ -6,19 +7,19 @@
   "bos_token_id": 0,
   "dropout": 0.0,
   "eos_token_id": 2,
-  "hidden_act": "quick_gelu",
-  "hidden_size": 768,
+  "hidden_act": "gelu",
+  "hidden_size": 1024,
   "initializer_factor": 1.0,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 77,
   "model_type": "clip_text_model",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "pad_token_id": 1,
-  "projection_dim": 768,
-  "torch_dtype": "float16",
-  "transformers_version": "4.33.3",
+  "projection_dim": 1024,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.0.dev0",
   "vocab_size": 49408
 }