Upload 6 files

config.json

@@ -1,25 +1,32 @@
 {
   "_name_or_path": "clip-aligned",
   "architectures": [
-    "CLIPTextModel"
+    "CLIPModel"
   ],
-  "attention_dropout": 0.0,
-  "bos_token_id": 0,
-  "dropout": 0.0,
-  "eos_token_id": 2,
-  "hidden_act": "quick_gelu",
-  "hidden_size": 768,
   "initializer_factor": 1.0,
-  "initializer_range": 0.02,
-  "intermediate_size": 3072,
-  "layer_norm_eps": 1e-05,
-  "max_position_embeddings": 512,
-  "model_type": "clip_text_model",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
-  "pad_token_id": 1,
+  "logit_scale_init_value": 2.6592,
+  "model_type": "clip",
   "projection_dim": 768,
+  "text_config": {
+    "dropout": 0.0,
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "max_position_embeddings": 512,
+    "model_type": "clip_text_model",
+    "num_attention_heads": 12,
+    "projection_dim": 768
+  },
   "torch_dtype": "float32",
   "transformers_version": "4.48.0",
+  "vision_config": {
+    "dropout": 0.0,
+    "hidden_size": 1024,
+    "intermediate_size": 4096,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768
+  },
   "vocab_size": 49408
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3062c0b2935dd8034dfea3a2e07513f236e874baad19b5c2e345e9bc069fd12
-size 493601488
+oid sha256:a71acf026dd6cb07cc4decf9ff3c7fc7d781f2b61e447b15ef11e52597e73172
+size 1711874036

tokenizer_config.json

@@ -24,7 +24,7 @@
   "eos_token": "<|endoftext|>",
   "errors": "replace",
   "extra_special_tokens": {},
-  "model_max_length": 512,
+  "model_max_length": 77,
   "pad_token": "<|endoftext|>",
   "tokenizer_class": "CLIPTokenizer",
   "unk_token": "<|endoftext|>"