fix?

config.json

@@ -190,7 +190,7 @@
     "image_size": 1024,
     "initializer_factor": 1.0,
     "initializer_range": 1e-10,
-    "intermediate_size": 192,
+    "intermediate_size": 768,
     "is_decoder": false,
     "is_encoder_decoder": false,
     "label2id": {
@@ -201,26 +201,26 @@
     "length_penalty": 1.0,
     "max_length": 20,
     "min_length": 0,
-    "mlp_dim": 192,
+    "mlp_dim": 384,
     "mlp_ratio": 4.0,
     "model_type": "",
     "no_repeat_ngram_size": 0,
-    "num_attention_heads": 12,
+    "num_attention_heads": 1,
     "num_beam_groups": 1,
     "num_beams": 1,
     "num_channels": 3,
     "num_hidden_layers": 12,
-    "num_pos_feats": 128,
+    "num_pos_feats": 16,
     "num_return_sequences": 1,
     "output_attentions": false,
-    "output_channels": 256,
+    "output_channels": 32,
     "output_hidden_states": false,
     "output_scores": false,
     "pad_token_id": null,
     "patch_size": 16,
     "prefix": null,
     "problem_type": null,
-    "projection_dim": 512,
+    "projection_dim": 64,
     "pruned_heads": {},
     "qkv_bias": true,
     "remove_invalid_values": false,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76919a7984a779bf380f4a826429a23ee893cf19a0677a2e8feadfcc5d3ca812
-size 9558581
+oid sha256:e990c8d7f134eb442a2dd3eb002674d758ccededbc620e85b882bf7fe23b0dc7
+size 9434869