Upload Pix2StructForConditionalGeneration

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "google/pix2struct-docvqa-large",
+  "architectures": [
+    "Pix2StructForConditionalGeneration"
+  ],
+  "decoder_start_token_id": 0,
+  "eos_token_id": 1,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "is_encoder_decoder": true,
+  "is_vqa": false,
+  "model_type": "pix2struct",
+  "pad_token_id": 0,
+  "text_config": {
+    "d_ff": 3968,
+    "encoder_hidden_size": 1536,
+    "hidden_size": 1536,
+    "initializer_range": 0.02,
+    "model_type": "pix2struct_text_model",
+    "num_heads": 24,
+    "num_layers": 18,
+    "vocab_size": 50348
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.0.dev0",
+  "vision_config": {
+    "d_ff": 3968,
+    "hidden_size": 1536,
+    "initializer_range": 0.02,
+    "layer_norm_bias": false,
+    "model_type": "pix2struct_vision_model",
+    "num_attention_heads": 24,
+    "num_channels": 3,
+    "num_hidden_layers": 18,
+    "patch_size": 16,
+    "projection_dim": 768
+  }
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "decoder_start_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.35.0.dev0",
+  "use_cache": false
+}

pytorch_model-00001-of-00002.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52d44471be6cfd2c278065ce0a10f8191079f7913dd89932589a7425b7ae0be7
+size 4987745089

pytorch_model-00002-of-00002.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0634c352de1f3e960b03868dafddf1dcbd6aed0b48936fce8e7bb84709d32f44
+size 358111342

pytorch_model.bin.index.json

@@ -0,0 +1,430 @@
+{
+  "metadata": {
+    "total_size": 5345707008
+  },
+  "weight_map": {
+    "decoder.embed_tokens.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.final_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "decoder.layer.0.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.self_attention.attention.relative_attention_bias.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.0.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.1.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.10.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.11.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.12.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.13.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.14.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.15.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.16.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00002-of-00002.bin",
+    "decoder.layer.17.mlp.DenseReluDense.wo.weight": "pytorch_model-00002-of-00002.bin",
+    "decoder.layer.17.mlp.layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "decoder.layer.17.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.17.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.2.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.3.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.4.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.5.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.6.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.7.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.8.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.encoder_decoder_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.encoder_decoder_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.encoder_decoder_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.encoder_decoder_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.encoder_decoder_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.mlp.DenseReluDense.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.mlp.DenseReluDense.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.mlp.DenseReluDense.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.mlp.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.self_attention.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.self_attention.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.self_attention.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.self_attention.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.layer.9.self_attention.layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "decoder.lm_head.weight": "pytorch_model-00002-of-00002.bin",
+    "encoder.embeddings.column_embedder.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.embeddings.patch_projection.bias": "pytorch_model-00001-of-00002.bin",
+    "encoder.embeddings.patch_projection.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.embeddings.row_embedder.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.0.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.1.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.10.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.11.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.12.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.13.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.14.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.15.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.16.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.17.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.2.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.3.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.4.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.5.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.6.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.7.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.8.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.attention.key.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.attention.output.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.attention.query.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.attention.value.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.mlp.wi_0.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.mlp.wi_1.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.mlp.wo.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.pre_attention_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.encoder.layer.9.pre_mlp_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "encoder.layernorm.weight": "pytorch_model-00001-of-00002.bin"
+  }
+}