Training in progress, step 500

config.json

@@ -10,10 +10,10 @@
   "num_query_tokens": 32,
   "qformer_config": {
     "_attn_implementation_autoset": true,
-    "encoder_hidden_size": 384,
-    "hidden_size": 384,
+    "encoder_hidden_size": 1024,
+    "hidden_size": 1024,
     "model_type": "blip_2_qformer",
-    "num_attention_heads": 6
+    "num_attention_heads": 16
   },
   "text_config": {
     "model_type": "opt"
@@ -23,10 +23,9 @@
   "use_decoder_only_language_model": true,
   "vision_config": {
     "_attn_implementation_autoset": true,
-    "hidden_size": 384,
+    "hidden_size": 1024,
     "image_size": 112,
     "model_type": "blip_2_vision_model",
-    "num_attention_heads": 6,
     "num_hidden_layers": 12,
     "patch_size": 16
   }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c5f7fa1c6880ab981751178504859163b0daafe8b5a5376d496f938a2d2e167
-size 914877216
+oid sha256:679d97ee0828ca76ac636976a4c2b451a8e2e6708cd83e1569b89b119e0245c4
+size 1918449256

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45f1638649a634ee184a2d85bbbca457c8f662deb9e36feb2a2deb0b0b97a83b
+oid sha256:02b3c3012ecb6b27ca0706706e413f3b12e943e407c72a7574cb80e0896494b1
 size 5368