tiny-random-z-image / text_encoder /openvino_model.xml
h0witended's picture
Add reference OpenVINO IRs
820f73a
Raw History Blame
92.2 kB
<?xml version="1.0"?>
<net name="Model0" version="11">
<layers>
<layer id="0" name="input_ids" type="Parameter" version="opset1">
<data shape="?,?" element_type="i64" />
<output>
<port id="0" precision="I64" names="input_ids">
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="1" name="attention_mask" type="Parameter" version="opset1">
<data shape="?,?" element_type="i64" />
<output>
<port id="0" precision="I64" names="attention_mask">
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="2" name="self.embed_tokens.weight" type="Const" version="opset1">
<data element_type="f32" shape="543, 32" offset="0" size="69504" />
<output>
<port id="0" precision="FP32" names="self.embed_tokens.weight">
<dim>543</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="3" name="__module.embed_tokens/aten::embedding/Convert" type="Convert" version="opset1">
<data destination_type="i32" />
<input>
<port id="0" precision="I64">
<dim>-1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="1" precision="I32">
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="4" name="__module.embed_tokens/aten::embedding/Constant" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="5" name="__module.embed_tokens/aten::embedding/Gather" type="Gather" version="opset8">
<data batch_dims="0" />
<input>
<port id="0" precision="FP32">
<dim>543</dim>
<dim>32</dim>
</port>
<port id="1" precision="I32">
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="2" precision="I32" />
</input>
<output>
<port id="3" precision="FP32" names="158,215,hidden_states.1,inputs_embeds">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="6" name="Constant_6543" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1" offset="69508" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="7" name="Constant_6541" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1" offset="69512" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="8" name="__module.layers.0.input_layernorm/aten::pow/Power" type="Power" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="216">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="9" name="Constant_449" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="69520" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="10" name="__module.layers.0.input_layernorm/aten::mean/ReduceMean" type="ReduceMean" version="opset1">
<data keep_dims="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="218,variance.1">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="11" name="Constant_6542" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1" offset="69528" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="12" name="__module.layers.0.input_layernorm/aten::add/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="219">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="13" name="__module.layers.0.input_layernorm/aten::rsqrt/Sqrt" type="Sqrt" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="14" name="__module.layers.0.input_layernorm/aten::rsqrt/Divide" type="Divide" version="opset1">
<data auto_broadcast="numpy" m_pythondiv="true" />
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="220">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="15" name="__module.layers.0.input_layernorm/aten::mul/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="221,222,223,hidden_states.3,hidden_states.5,hidden_states.7">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="16" name="self.layers.0.self_attn.q_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="32, 32" offset="69532" size="4096" />
<output>
<port id="0" precision="FP32" names="self.layers.0.self_attn.q_proj.weight">
<dim>32</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="17" name="__module.layers.0.self_attn.q_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>32</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="236">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="18" name="Constant_6659" type="Const" version="opset1">
<data element_type="i64" shape="4" offset="73632" size="32" />
<output>
<port id="0" precision="I64">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="19" name="__module.layers.0.self_attn/aten::view/Reshape" type="Reshape" version="opset1">
<data special_zero="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="I64">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="238,240,hidden_states.11,hidden_states.9">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="20" name="Constant_6546" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1, 1" offset="69508" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="21" name="Constant_6544" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1, 1" offset="69512" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="22" name="__module.layers.0.self_attn.q_norm/aten::pow/Power" type="Power" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="241">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="23" name="Constant_527" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="69520" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="24" name="__module.layers.0.self_attn.q_norm/aten::mean/ReduceMean" type="ReduceMean" version="opset1">
<data keep_dims="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="243,variance.3">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="25" name="Constant_6545" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1, 1" offset="69528" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="26" name="__module.layers.0.self_attn.q_norm/aten::add/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="244">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="27" name="__module.layers.0.self_attn.q_norm/aten::rsqrt/Sqrt" type="Sqrt" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="28" name="__module.layers.0.self_attn.q_norm/aten::rsqrt/Divide" type="Divide" version="opset1">
<data auto_broadcast="numpy" m_pythondiv="true" />
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="245">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="29" name="__module.layers.0.self_attn.q_norm/aten::mul/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="246,247,248,hidden_states.13,hidden_states.15">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="30" name="__module.layers.0.self_attn/aten::transpose/Constant" type="Const" version="opset1">
<data element_type="i32" shape="4" offset="73664" size="16" />
<output>
<port id="0" precision="I32">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="31" name="__module.layers.0.self_attn/aten::transpose/Transpose" type="Transpose" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="I32">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="249,q.1">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="32" name="__module.rotary_emb/aten::to/Convert_2" type="Const" version="opset1">
<data element_type="f32" shape="1, 8, 1" offset="73680" size="32" />
<output>
<port id="0" precision="FP32" names="182,inv_freq_expanded">
<dim>1</dim>
<dim>8</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="33" name="aten::arange/Constant" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="34" name="ShapeOf_6637" type="ShapeOf" version="opset3">
<data output_type="i64" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="1" precision="I64">
<dim>3</dim>
</port>
</output>
</layer>
<layer id="35" name="Constant_6638" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73712" size="8" />
<output>
<port id="0" precision="I64" />
</output>
</layer>
<layer id="36" name="Constant_6639" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" />
</output>
</layer>
<layer id="37" name="Gather_6640" type="Gather" version="opset8">
<data batch_dims="0" />
<input>
<port id="0" precision="I64">
<dim>3</dim>
</port>
<port id="1" precision="I64" />
<port id="2" precision="I64" />
</input>
<output>
<port id="3" precision="I64" names="12,32,35,37,46" />
</output>
</layer>
<layer id="38" name="aten::arange/Constant_1" type="Const" version="opset1">
<data element_type="i32" shape="" offset="73728" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="39" name="aten::arange/Range" type="Range" version="opset4">
<data output_type="i64" />
<input>
<port id="0" precision="I32" />
<port id="1" precision="I64" />
<port id="2" precision="I32" />
</input>
<output>
<port id="3" precision="I64" names="19,22,78,81,kv_indices,position_ids.1">
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="40" name="23" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" names="23" />
</output>
</layer>
<layer id="41" name="aten::unsqueeze/Unsqueeze" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="105,178,24,position_ids">
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="42" name="167" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73712" size="8" />
<output>
<port id="0" precision="I64" names="167" />
</output>
</layer>
<layer id="43" name="__module.rotary_emb/aten::unsqueeze/Unsqueeze_2" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="107,179,180">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="44" name="__module.rotary_emb/aten::to/Convert_3" type="Convert" version="opset1">
<data destination_type="f32" />
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="1" precision="FP32" names="181,183,position_ids_expanded,position_ids_expanded.1">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="45" name="__module.rotary_emb/aten::matmul/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="false" />
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>8</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="184">
<dim>1</dim>
<dim>8</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="46" name="__module.rotary_emb/aten::transpose/Constant" type="Const" version="opset1">
<data element_type="i32" shape="3" offset="73732" size="12" />
<output>
<port id="0" precision="I32">
<dim>3</dim>
</port>
</output>
</layer>
<layer id="47" name="__module.rotary_emb/aten::transpose/Transpose" type="Transpose" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>8</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I32">
<dim>3</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="185,freqs">
<dim>1</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="48" name="__module.rotary_emb/aten::cat/Concat" type="Concat" version="opset1">
<data axis="-1" />
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="187,emb">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="49" name="__module.rotary_emb/aten::cos/Cos" type="Cos" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="1" precision="FP32" names="188,189,192,cos.1,cos.3">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="50" name="209" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73712" size="8" />
<output>
<port id="0" precision="I64" names="209" />
</output>
</layer>
<layer id="51" name="__module.layers.0.self_attn/aten::unsqueeze/Unsqueeze" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="FP32" names="270,cos.5">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="52" name="__module.layers.0.self_attn/aten::mul/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="272">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="53" name="Constant_850" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73744" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="54" name="Constant_852" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73752" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="55" name="Constant_854" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="56" name="Constant_806" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73760" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="57" name="__module.layers.0.self_attn/aten::slice/Slice" type="Slice" version="opset8">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
<port id="2" precision="I64">
<dim>1</dim>
</port>
<port id="3" precision="I64">
<dim>1</dim>
</port>
<port id="4" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="5" precision="FP32" names="282,x2.1">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="58" name="__module.layers.0.self_attn/aten::neg/Constant" type="Const" version="opset1">
<data element_type="i32" shape="" offset="73768" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="59" name="__module.layers.0.self_attn/aten::neg/ConvertLike" type="Convert" version="opset1">
<data destination_type="f32" />
<input>
<port id="0" precision="I32" />
</input>
<output>
<port id="1" precision="FP32" />
</output>
</layer>
<layer id="60" name="__module.layers.0.self_attn/aten::neg/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
<port id="1" precision="FP32" />
</input>
<output>
<port id="2" precision="FP32" names="283">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="61" name="Constant_747" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73720" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="62" name="Constant_791" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73744" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="63" name="Constant_793" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="64" name="Constant_745" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73760" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="65" name="__module.layers.0.self_attn/aten::slice/Slice_1" type="Slice" version="opset8">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
<port id="2" precision="I64">
<dim>1</dim>
</port>
<port id="3" precision="I64">
<dim>1</dim>
</port>
<port id="4" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="5" precision="FP32" names="277,x1.1">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="66" name="__module.layers.0.self_attn/aten::cat/Concat" type="Concat" version="opset1">
<data axis="-1" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="285">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="67" name="__module.rotary_emb/aten::sin/Sin" type="Sin" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="1" precision="FP32" names="190,191,193,sin.1,sin.3">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="68" name="__module.layers.0.self_attn/aten::unsqueeze/Unsqueeze_1" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="FP32" names="271,sin.5">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="69" name="__module.layers.0.self_attn/aten::mul/Multiply_1" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="286">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="70" name="__module.layers.0.self_attn/aten::add/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="287,query.1">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="71" name="self.layers.0.self_attn.k_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="32, 32" offset="73772" size="4096" />
<output>
<port id="0" precision="FP32" names="self.layers.0.self_attn.k_proj.weight">
<dim>32</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="72" name="__module.layers.0.self_attn.k_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>32</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="251">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="73" name="Constant_6660" type="Const" version="opset1">
<data element_type="i64" shape="4" offset="73632" size="32" />
<output>
<port id="0" precision="I64">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="74" name="__module.layers.0.self_attn/aten::view/Reshape_1" type="Reshape" version="opset1">
<data special_zero="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="I64">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="253,255,hidden_states.17,hidden_states.19">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="75" name="Constant_6549" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1, 1" offset="69508" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="76" name="Constant_6547" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1, 1" offset="69512" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="77" name="__module.layers.0.self_attn.k_norm/aten::pow/Power" type="Power" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="256">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="78" name="Constant_624" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="69520" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="79" name="__module.layers.0.self_attn.k_norm/aten::mean/ReduceMean" type="ReduceMean" version="opset1">
<data keep_dims="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="258,variance.5">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="80" name="Constant_6548" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1, 1" offset="69528" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="81" name="__module.layers.0.self_attn.k_norm/aten::add/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="259">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="82" name="__module.layers.0.self_attn.k_norm/aten::rsqrt/Sqrt" type="Sqrt" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="83" name="__module.layers.0.self_attn.k_norm/aten::rsqrt/Divide" type="Divide" version="opset1">
<data auto_broadcast="numpy" m_pythondiv="true" />
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="260">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="84" name="__module.layers.0.self_attn.k_norm/aten::mul/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="261,262,263,hidden_states.21,hidden_states.23">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="85" name="__module.layers.0.self_attn/aten::transpose/Constant_1" type="Const" version="opset1">
<data element_type="i32" shape="4" offset="73664" size="16" />
<output>
<port id="0" precision="I32">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="86" name="__module.layers.0.self_attn/aten::transpose/Transpose_1" type="Transpose" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="I32">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="264,k.1">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="87" name="__module.layers.0.self_attn/aten::mul/Multiply_2" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="288">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="88" name="Constant_992" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73744" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="89" name="Constant_994" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73752" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="90" name="Constant_996" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="91" name="Constant_948" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73760" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="92" name="__module.layers.0.self_attn/aten::slice/Slice_2" type="Slice" version="opset8">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
<port id="2" precision="I64">
<dim>1</dim>
</port>
<port id="3" precision="I64">
<dim>1</dim>
</port>
<port id="4" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="5" precision="FP32" names="298,x2.3">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="93" name="__module.layers.0.self_attn/aten::neg/Constant_1" type="Const" version="opset1">
<data element_type="i32" shape="" offset="73768" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="94" name="__module.layers.0.self_attn/aten::neg/ConvertLike_1" type="Convert" version="opset1">
<data destination_type="f32" />
<input>
<port id="0" precision="I32" />
</input>
<output>
<port id="1" precision="FP32" />
</output>
</layer>
<layer id="95" name="__module.layers.0.self_attn/aten::neg/Multiply_1" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
<port id="1" precision="FP32" />
</input>
<output>
<port id="2" precision="FP32" names="299">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="96" name="Constant_889" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73720" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="97" name="Constant_933" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73744" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="98" name="Constant_935" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="99" name="Constant_887" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73760" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="100" name="__module.layers.0.self_attn/aten::slice/Slice_3" type="Slice" version="opset8">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
<port id="2" precision="I64">
<dim>1</dim>
</port>
<port id="3" precision="I64">
<dim>1</dim>
</port>
<port id="4" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="5" precision="FP32" names="293,x1.3">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</output>
</layer>
<layer id="101" name="__module.layers.0.self_attn/aten::cat/Concat_1" type="Concat" version="opset1">
<data axis="-1" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>8</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="301">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="102" name="__module.layers.0.self_attn/aten::mul/Multiply_3" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="302">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="103" name="__module.layers.0.self_attn/aten::add/Add_1" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="303,hidden_states.25">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="104" name="self.layers.0.self_attn.v_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="32, 32" offset="77868" size="4096" />
<output>
<port id="0" precision="FP32" names="self.layers.0.self_attn.v_proj.weight">
<dim>32</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="105" name="__module.layers.0.self_attn.v_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>32</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="266">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="106" name="Constant_6661" type="Const" version="opset1">
<data element_type="i64" shape="4" offset="73632" size="32" />
<output>
<port id="0" precision="I64">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="107" name="__module.layers.0.self_attn/aten::view/Reshape_2" type="Reshape" version="opset1">
<data special_zero="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="I64">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="268">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="108" name="__module.layers.0.self_attn/aten::transpose/Constant_2" type="Const" version="opset1">
<data element_type="i32" shape="4" offset="73664" size="16" />
<output>
<port id="0" precision="I32">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="109" name="__module.layers.0.self_attn/aten::transpose/Transpose_2" type="Transpose" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="I32">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="269,hidden_states.27">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="110" name="aten::new_ones/Broadcast" type="Const" version="opset1">
<data element_type="boolean" shape="" offset="81964" size="1" />
<output>
<port id="0" precision="BOOL" names="120,result.1" />
</output>
</layer>
<layer id="111" name="108" type="Const" version="opset1">
<data element_type="i64" shape="" offset="81968" size="8" />
<output>
<port id="0" precision="I64" names="108" />
</output>
</layer>
<layer id="112" name="aten::unsqueeze/Unsqueeze_3" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="109,114,kv_idx">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="113" name="aten::arange/Constant_4" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="114" name="Constant_6647" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="115" name="Reshape_6648" type="Reshape" version="opset1">
<data special_zero="false" />
<input>
<port id="0" precision="I64" />
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="116" name="aten::size/Constant_3" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="117" name="aten::size/Gather_3" type="Squeeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>1</dim>
</port>
<port id="1" precision="I32" />
</input>
<output>
<port id="2" precision="I64" names="55,57" />
</output>
</layer>
<layer id="118" name="aten::arange/Constant_6" type="Const" version="opset1">
<data element_type="i32" shape="" offset="73728" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="119" name="aten::arange/Range_3" type="Range" version="opset4">
<data output_type="i64" />
<input>
<port id="0" precision="I32" />
<port id="1" precision="I64" />
<port id="2" precision="I32" />
</input>
<output>
<port id="3" precision="I64" names="71">
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="120" name="48" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" names="48" />
</output>
</layer>
<layer id="121" name="aten::arange/Constant_5" type="Const" version="opset1">
<data element_type="i32" shape="" offset="73728" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="122" name="aten::arange/Range_2" type="Range" version="opset4">
<data output_type="i64" />
<input>
<port id="0" precision="I64" />
<port id="1" precision="I64" />
<port id="2" precision="I32" />
</input>
<output>
<port id="3" precision="I64" names="53,cache_position">
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="123" name="60" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" names="60" />
</output>
</layer>
<layer id="124" name="59" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" names="59" />
</output>
</layer>
<layer id="125" name="aten::select/Gather" type="Gather" version="opset8">
<data batch_dims="0" />
<input>
<port id="0" precision="I64">
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
<port id="2" precision="I64" />
</input>
<output>
<port id="3" precision="I64" names="61,q_offset" />
</output>
</layer>
<layer id="126" name="aten::add/Add_3" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="I64">
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="73,q_indices">
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="127" name="93" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" names="93" />
</output>
</layer>
<layer id="128" name="aten::unsqueeze/Unsqueeze_4" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="94">
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="129" name="95" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73712" size="8" />
<output>
<port id="0" precision="I64" names="95" />
</output>
</layer>
<layer id="130" name="aten::unsqueeze/Unsqueeze_5" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="101,96">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="131" name="102" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73760" size="8" />
<output>
<port id="0" precision="I64" names="102" />
</output>
</layer>
<layer id="132" name="aten::unsqueeze/Unsqueeze_6" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="103,q_idx">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="133" name="aten::le/LessEqual" type="LessEqual" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL" names="121,129">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="134" name="aten::__and__/BitwiseAnd" type="BitwiseAnd" version="opset13">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="BOOL" />
<port id="1" precision="BOOL">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL" names="130,result">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="135" name="aten::to/Convert" type="Convert" version="opset1">
<data destination_type="boolean" />
<input>
<port id="0" precision="I64">
<dim>-1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="1" precision="BOOL" names="30,attention_mask.3">
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="136" name="Constant_5836" type="Const" version="opset1">
<data element_type="i64" shape="2" offset="81976" size="16" />
<output>
<port id="0" precision="I64">
<dim>2</dim>
</port>
</output>
</layer>
<layer id="137" name="Constant_5837" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" />
</output>
</layer>
<layer id="138" name="Gather_5838" type="Gather" version="opset8">
<data batch_dims="0" />
<input>
<port id="0" precision="I64">
<dim>3</dim>
</port>
<port id="1" precision="I64">
<dim>2</dim>
</port>
<port id="2" precision="I64" />
</input>
<output>
<port id="3" precision="I64">
<dim>2</dim>
</port>
</output>
</layer>
<layer id="139" name="Convert_6630" type="Convert" version="opset1">
<data destination_type="i32" />
<input>
<port id="0" precision="I64">
<dim>2</dim>
</port>
</input>
<output>
<port id="1" precision="I32">
<dim>2</dim>
</port>
</output>
</layer>
<layer id="140" name="Constant_1943" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="141" name="aten::index/ReduceProd" type="ReduceProd" version="opset1">
<data keep_dims="true" />
<input>
<port id="0" precision="I32">
<dim>2</dim>
</port>
<port id="1" precision="I32" />
</input>
<output>
<port id="2" precision="I32">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="142" name="Constant_1945" type="Const" version="opset1">
<data element_type="i32" shape="1" offset="73768" size="4" />
<output>
<port id="0" precision="I32">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="143" name="aten::index/Concat" type="Concat" version="opset1">
<data axis="0" />
<input>
<port id="0" precision="I32">
<dim>1</dim>
</port>
<port id="1" precision="I32">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="I32">
<dim>2</dim>
</port>
</output>
</layer>
<layer id="144" name="aten::index/Reshape" type="Reshape" version="opset1">
<data special_zero="true" />
<input>
<port id="0" precision="BOOL">
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I32">
<dim>2</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL">
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="145" name="aten::index/Convert" type="Convert" version="opset1">
<data destination_type="i32" />
<input>
<port id="0" precision="I64">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="1" precision="I32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="146" name="aten::arange/Constant_7" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="147" name="Constant_6642" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" />
</output>
</layer>
<layer id="148" name="Constant_6643" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73720" size="8" />
<output>
<port id="0" precision="I64" />
</output>
</layer>
<layer id="149" name="Gather_6644" type="Gather" version="opset8">
<data batch_dims="0" />
<input>
<port id="0" precision="I64">
<dim>3</dim>
</port>
<port id="1" precision="I64" />
<port id="2" precision="I64" />
</input>
<output>
<port id="3" precision="I64" names="40,42" />
</output>
</layer>
<layer id="150" name="aten::arange/Constant_8" type="Const" version="opset1">
<data element_type="i32" shape="" offset="73728" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="151" name="aten::arange/Range_4" type="Range" version="opset4">
<data output_type="i64" />
<input>
<port id="0" precision="I32" />
<port id="1" precision="I64" />
<port id="2" precision="I32" />
</input>
<output>
<port id="3" precision="I64" names="66,86,batch_indices">
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="152" name="87" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73712" size="8" />
<output>
<port id="0" precision="I64" names="87" />
</output>
</layer>
<layer id="153" name="aten::unsqueeze/Unsqueeze_7" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>-1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="88">
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="154" name="89" type="Const" version="opset1">
<data element_type="i64" shape="" offset="81968" size="8" />
<output>
<port id="0" precision="I64" names="89" />
</output>
</layer>
<layer id="155" name="aten::unsqueeze/Unsqueeze_8" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>-1</dim>
<dim>1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="90">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="156" name="91" type="Const" version="opset1">
<data element_type="i64" shape="" offset="73760" size="8" />
<output>
<port id="0" precision="I64" names="91" />
</output>
</layer>
<layer id="157" name="aten::unsqueeze/Unsqueeze_9" type="Unsqueeze" version="opset1">
<input>
<port id="0" precision="I64">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
<port id="1" precision="I64" />
</input>
<output>
<port id="2" precision="I64" names="92,batch_idx">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="158" name="aten::index/Convert_3" type="Convert" version="opset1">
<data destination_type="i32" />
<input>
<port id="0" precision="I64">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="1" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="159" name="aten::index/Constant" type="Const" version="opset1">
<data element_type="i32" shape="" offset="69504" size="4" />
<output>
<port id="0" precision="I32" />
</output>
</layer>
<layer id="160" name="aten::index/Split" type="Split" version="opset1">
<data num_splits="2" />
<input>
<port id="0" precision="I32">
<dim>2</dim>
</port>
<port id="1" precision="I32" />
</input>
<output>
<port id="2" precision="I32">
<dim>1</dim>
</port>
<port id="3" precision="I32">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="161" name="aten::index/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
<port id="1" precision="I32">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="162" name="aten::index/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="I32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="163" name="aten::index/Gather" type="Gather" version="opset8">
<data batch_dims="0" />
<input>
<port id="0" precision="BOOL">
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
<port id="2" precision="I32" />
</input>
<output>
<port id="3" precision="BOOL">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="164" name="Constant_3305" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="69520" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="165" name="aten::index/Reshape_8" type="Reshape" version="opset1">
<data special_zero="false" />
<input>
<port id="0" precision="BOOL">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL">
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="166" name="aten::index/ShapeOf_5" type="ShapeOf" version="opset3">
<data output_type="i32" />
<input>
<port id="0" precision="I32">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="1" precision="I32">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="167" name="aten::index/Reshape_12" type="Reshape" version="opset1">
<data special_zero="false" />
<input>
<port id="0" precision="BOOL">
<dim>-1</dim>
</port>
<port id="1" precision="I32">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL" names="132,140">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="168" name="aten::__and__/BitwiseAnd_1" type="BitwiseAnd" version="opset13">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="BOOL">
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="BOOL">
<dim>-1</dim>
<dim>1</dim>
<dim>1</dim>
<dim>-1</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL" names="141,attention_mask.5">
<dim>-1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="169" name="Constant_6649" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="170" name="Reshape_6650" type="Reshape" version="opset1">
<data special_zero="false" />
<input>
<port id="0" precision="I64" />
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="171" name="Constant_5817" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="73712" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="172" name="prim::ListConstruct/SequenceMark" type="Concat" version="opset1">
<data axis="0" />
<input>
<port id="0" precision="I64">
<dim>1</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
<port id="2" precision="I64">
<dim>1</dim>
</port>
<port id="3" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="4" precision="I64">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="173" name="aten::expand/Broadcast" type="Broadcast" version="opset3">
<data mode="bidirectional" />
<input>
<port id="0" precision="BOOL">
<dim>-1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="1" precision="I64">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="BOOL" names="145,attention_mask_1">
<dim>-1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
</output>
</layer>
<layer id="174" name="__module.layers.0.self_attn/aten::scaled_dot_product_attention/ConvertLike" type="Const" version="opset1">
<data element_type="f32" shape="" offset="81992" size="4" />
<output>
<port id="0" precision="FP32" />
</output>
</layer>
<layer id="175" name="__module.layers.0.self_attn/aten::scaled_dot_product_attention/ScaledDotProductAttention" type="ScaledDotProductAttention" version="opset13">
<data causal="false" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="2" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="3" precision="BOOL">
<dim>-1</dim>
<dim>1</dim>
<dim>-1</dim>
<dim>-1</dim>
</port>
<port id="4" precision="FP32" />
</input>
<output>
<port id="5" precision="FP32" names="304,attn_output.1">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="176" name="__module.layers.0.self_attn/aten::transpose/Constant_3" type="Const" version="opset1">
<data element_type="i32" shape="4" offset="73664" size="16" />
<output>
<port id="0" precision="I32">
<dim>4</dim>
</port>
</output>
</layer>
<layer id="177" name="__module.layers.0.self_attn/aten::transpose/Transpose_3" type="Transpose" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>2</dim>
<dim>-1</dim>
<dim>16</dim>
</port>
<port id="1" precision="I32">
<dim>4</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="305">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
</output>
</layer>
<layer id="178" name="Constant_6662" type="Const" version="opset1">
<data element_type="i64" shape="3" offset="82000" size="24" />
<output>
<port id="0" precision="I64">
<dim>3</dim>
</port>
</output>
</layer>
<layer id="179" name="__module.layers.0.self_attn/aten::reshape/Reshape" type="Reshape" version="opset1">
<data special_zero="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>2</dim>
<dim>16</dim>
</port>
<port id="1" precision="I64">
<dim>3</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="308">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="180" name="self.layers.0.self_attn.o_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="32, 32" offset="82024" size="4096" />
<output>
<port id="0" precision="FP32" names="self.layers.0.self_attn.o_proj.weight">
<dim>32</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="181" name="__module.layers.0.self_attn.o_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>32</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="311,hidden_states.29">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="182" name="__module.layers.0/aten::add/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="312,314,hidden_states.31,hidden_states.33">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="183" name="Constant_6552" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1" offset="69508" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="184" name="Constant_6550" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1" offset="69512" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="185" name="__module.layers.0.post_attention_layernorm/aten::pow/Power" type="Power" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="315">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="186" name="Constant_1090" type="Const" version="opset1">
<data element_type="i64" shape="1" offset="69520" size="8" />
<output>
<port id="0" precision="I64">
<dim>1</dim>
</port>
</output>
</layer>
<layer id="187" name="__module.layers.0.post_attention_layernorm/aten::mean/ReduceMean" type="ReduceMean" version="opset1">
<data keep_dims="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="I64">
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="317,variance.7">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="188" name="Constant_6551" type="Const" version="opset1">
<data element_type="f32" shape="1, 1, 1" offset="69528" size="4" />
<output>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="189" name="__module.layers.0.post_attention_layernorm/aten::add/Add" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="318">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="190" name="__module.layers.0.post_attention_layernorm/aten::rsqrt/Sqrt" type="Sqrt" version="opset1">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="191" name="__module.layers.0.post_attention_layernorm/aten::rsqrt/Divide" type="Divide" version="opset1">
<data auto_broadcast="numpy" m_pythondiv="true" />
<input>
<port id="0" precision="FP32">
<dim>1</dim>
<dim>1</dim>
<dim>1</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="319">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</output>
</layer>
<layer id="192" name="__module.layers.0.post_attention_layernorm/aten::mul/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>1</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="320,321,322,hidden_states.35,hidden_states.37,input.3">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="193" name="self.layers.0.mlp.gate_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="64, 32" offset="86120" size="8192" />
<output>
<port id="0" precision="FP32" names="self.layers.0.mlp.gate_proj.weight">
<dim>64</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="194" name="__module.layers.0.mlp.gate_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>64</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="331,input.5">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
</output>
</layer>
<layer id="195" name="__module.layers.0.mlp.act_fn/aten::silu/Swish" type="Swish" version="opset4">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
</input>
<output>
<port id="1" precision="FP32" names="332">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
</output>
</layer>
<layer id="196" name="self.layers.0.mlp.up_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="64, 32" offset="94312" size="8192" />
<output>
<port id="0" precision="FP32" names="self.layers.0.mlp.up_proj.weight">
<dim>64</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="197" name="__module.layers.0.mlp.up_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>64</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="334">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
</output>
</layer>
<layer id="198" name="__module.layers.0.mlp/aten::mul/Multiply" type="Multiply" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="335,input.7">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
</output>
</layer>
<layer id="199" name="self.layers.0.mlp.down_proj.weight" type="Const" version="opset1">
<data element_type="f32" shape="32, 64" offset="102504" size="8192" />
<output>
<port id="0" precision="FP32" names="self.layers.0.mlp.down_proj.weight">
<dim>32</dim>
<dim>64</dim>
</port>
</output>
</layer>
<layer id="200" name="__module.layers.0.mlp.down_proj/aten::linear/MatMul" type="MatMul" version="opset1">
<data transpose_a="false" transpose_b="true" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>64</dim>
</port>
<port id="1" precision="FP32">
<dim>32</dim>
<dim>64</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="337,hidden_states.39">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="201" name="__module.layers.1.input_layernorm/aten::to/Convert" type="Add" version="opset1">
<data auto_broadcast="numpy" />
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
<port id="1" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</input>
<output>
<port id="2" precision="FP32" names="last_hidden_state">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</output>
</layer>
<layer id="202" name="Result_1992" type="Result" version="opset1" output_names="last_hidden_state">
<input>
<port id="0" precision="FP32">
<dim>-1</dim>
<dim>-1</dim>
<dim>32</dim>
</port>
</input>
</layer>
</layers>
<edges>
<edge from-layer="0" from-port="0" to-layer="3" to-port="0" />
<edge from-layer="1" from-port="0" to-layer="135" to-port="0" />
<edge from-layer="2" from-port="0" to-layer="5" to-port="0" />
<edge from-layer="3" from-port="1" to-layer="5" to-port="1" />
<edge from-layer="4" from-port="0" to-layer="5" to-port="2" />
<edge from-layer="5" from-port="3" to-layer="8" to-port="0" />
<edge from-layer="5" from-port="3" to-layer="15" to-port="0" />
<edge from-layer="5" from-port="3" to-layer="182" to-port="0" />
<edge from-layer="6" from-port="0" to-layer="14" to-port="0" />
<edge from-layer="7" from-port="0" to-layer="8" to-port="1" />
<edge from-layer="8" from-port="2" to-layer="10" to-port="0" />
<edge from-layer="9" from-port="0" to-layer="10" to-port="1" />
<edge from-layer="10" from-port="2" to-layer="12" to-port="0" />
<edge from-layer="11" from-port="0" to-layer="12" to-port="1" />
<edge from-layer="12" from-port="2" to-layer="13" to-port="0" />
<edge from-layer="13" from-port="1" to-layer="14" to-port="1" />
<edge from-layer="14" from-port="2" to-layer="15" to-port="1" />
<edge from-layer="15" from-port="2" to-layer="72" to-port="0" />
<edge from-layer="15" from-port="2" to-layer="105" to-port="0" />
<edge from-layer="15" from-port="2" to-layer="17" to-port="0" />
<edge from-layer="15" from-port="2" to-layer="34" to-port="0" />
<edge from-layer="16" from-port="0" to-layer="17" to-port="1" />
<edge from-layer="17" from-port="2" to-layer="19" to-port="0" />
<edge from-layer="18" from-port="0" to-layer="19" to-port="1" />
<edge from-layer="19" from-port="2" to-layer="22" to-port="0" />
<edge from-layer="19" from-port="2" to-layer="29" to-port="0" />
<edge from-layer="20" from-port="0" to-layer="28" to-port="0" />
<edge from-layer="21" from-port="0" to-layer="22" to-port="1" />
<edge from-layer="22" from-port="2" to-layer="24" to-port="0" />
<edge from-layer="23" from-port="0" to-layer="24" to-port="1" />
<edge from-layer="24" from-port="2" to-layer="26" to-port="0" />
<edge from-layer="25" from-port="0" to-layer="26" to-port="1" />
<edge from-layer="26" from-port="2" to-layer="27" to-port="0" />
<edge from-layer="27" from-port="1" to-layer="28" to-port="1" />
<edge from-layer="28" from-port="2" to-layer="29" to-port="1" />
<edge from-layer="29" from-port="2" to-layer="31" to-port="0" />
<edge from-layer="30" from-port="0" to-layer="31" to-port="1" />
<edge from-layer="31" from-port="2" to-layer="65" to-port="0" />
<edge from-layer="31" from-port="2" to-layer="57" to-port="0" />
<edge from-layer="31" from-port="2" to-layer="52" to-port="0" />
<edge from-layer="32" from-port="0" to-layer="45" to-port="0" />
<edge from-layer="33" from-port="0" to-layer="39" to-port="0" />
<edge from-layer="34" from-port="1" to-layer="37" to-port="0" />
<edge from-layer="34" from-port="1" to-layer="138" to-port="0" />
<edge from-layer="34" from-port="1" to-layer="149" to-port="0" />
<edge from-layer="35" from-port="0" to-layer="37" to-port="1" />
<edge from-layer="36" from-port="0" to-layer="37" to-port="2" />
<edge from-layer="37" from-port="3" to-layer="122" to-port="1" />
<edge from-layer="37" from-port="3" to-layer="115" to-port="0" />
<edge from-layer="37" from-port="3" to-layer="39" to-port="1" />
<edge from-layer="38" from-port="0" to-layer="39" to-port="2" />
<edge from-layer="39" from-port="3" to-layer="41" to-port="0" />
<edge from-layer="40" from-port="0" to-layer="41" to-port="1" />
<edge from-layer="41" from-port="2" to-layer="43" to-port="0" />
<edge from-layer="42" from-port="0" to-layer="43" to-port="1" />
<edge from-layer="43" from-port="2" to-layer="44" to-port="0" />
<edge from-layer="43" from-port="2" to-layer="112" to-port="0" />
<edge from-layer="44" from-port="1" to-layer="45" to-port="1" />
<edge from-layer="45" from-port="2" to-layer="47" to-port="0" />
<edge from-layer="46" from-port="0" to-layer="47" to-port="1" />
<edge from-layer="47" from-port="2" to-layer="48" to-port="0" />
<edge from-layer="47" from-port="2" to-layer="48" to-port="1" />
<edge from-layer="48" from-port="2" to-layer="49" to-port="0" />
<edge from-layer="48" from-port="2" to-layer="67" to-port="0" />
<edge from-layer="49" from-port="1" to-layer="51" to-port="0" />
<edge from-layer="50" from-port="0" to-layer="51" to-port="1" />
<edge from-layer="50" from-port="0" to-layer="68" to-port="1" />
<edge from-layer="51" from-port="2" to-layer="52" to-port="1" />
<edge from-layer="51" from-port="2" to-layer="87" to-port="1" />
<edge from-layer="52" from-port="2" to-layer="70" to-port="0" />
<edge from-layer="53" from-port="0" to-layer="57" to-port="1" />
<edge from-layer="54" from-port="0" to-layer="57" to-port="2" />
<edge from-layer="55" from-port="0" to-layer="57" to-port="3" />
<edge from-layer="56" from-port="0" to-layer="57" to-port="4" />
<edge from-layer="57" from-port="5" to-layer="60" to-port="0" />
<edge from-layer="58" from-port="0" to-layer="59" to-port="0" />
<edge from-layer="59" from-port="1" to-layer="60" to-port="1" />
<edge from-layer="60" from-port="2" to-layer="66" to-port="0" />
<edge from-layer="61" from-port="0" to-layer="65" to-port="1" />
<edge from-layer="62" from-port="0" to-layer="65" to-port="2" />
<edge from-layer="63" from-port="0" to-layer="65" to-port="3" />
<edge from-layer="64" from-port="0" to-layer="65" to-port="4" />
<edge from-layer="65" from-port="5" to-layer="66" to-port="1" />
<edge from-layer="66" from-port="2" to-layer="69" to-port="0" />
<edge from-layer="67" from-port="1" to-layer="68" to-port="0" />
<edge from-layer="68" from-port="2" to-layer="69" to-port="1" />
<edge from-layer="68" from-port="2" to-layer="102" to-port="1" />
<edge from-layer="69" from-port="2" to-layer="70" to-port="1" />
<edge from-layer="70" from-port="2" to-layer="175" to-port="0" />
<edge from-layer="71" from-port="0" to-layer="72" to-port="1" />
<edge from-layer="72" from-port="2" to-layer="74" to-port="0" />
<edge from-layer="73" from-port="0" to-layer="74" to-port="1" />
<edge from-layer="74" from-port="2" to-layer="77" to-port="0" />
<edge from-layer="74" from-port="2" to-layer="84" to-port="0" />
<edge from-layer="75" from-port="0" to-layer="83" to-port="0" />
<edge from-layer="76" from-port="0" to-layer="77" to-port="1" />
<edge from-layer="77" from-port="2" to-layer="79" to-port="0" />
<edge from-layer="78" from-port="0" to-layer="79" to-port="1" />
<edge from-layer="79" from-port="2" to-layer="81" to-port="0" />
<edge from-layer="80" from-port="0" to-layer="81" to-port="1" />
<edge from-layer="81" from-port="2" to-layer="82" to-port="0" />
<edge from-layer="82" from-port="1" to-layer="83" to-port="1" />
<edge from-layer="83" from-port="2" to-layer="84" to-port="1" />
<edge from-layer="84" from-port="2" to-layer="86" to-port="0" />
<edge from-layer="85" from-port="0" to-layer="86" to-port="1" />
<edge from-layer="86" from-port="2" to-layer="100" to-port="0" />
<edge from-layer="86" from-port="2" to-layer="87" to-port="0" />
<edge from-layer="86" from-port="2" to-layer="92" to-port="0" />
<edge from-layer="87" from-port="2" to-layer="103" to-port="0" />
<edge from-layer="88" from-port="0" to-layer="92" to-port="1" />
<edge from-layer="89" from-port="0" to-layer="92" to-port="2" />
<edge from-layer="90" from-port="0" to-layer="92" to-port="3" />
<edge from-layer="91" from-port="0" to-layer="92" to-port="4" />
<edge from-layer="92" from-port="5" to-layer="95" to-port="0" />
<edge from-layer="93" from-port="0" to-layer="94" to-port="0" />
<edge from-layer="94" from-port="1" to-layer="95" to-port="1" />
<edge from-layer="95" from-port="2" to-layer="101" to-port="0" />
<edge from-layer="96" from-port="0" to-layer="100" to-port="1" />
<edge from-layer="97" from-port="0" to-layer="100" to-port="2" />
<edge from-layer="98" from-port="0" to-layer="100" to-port="3" />
<edge from-layer="99" from-port="0" to-layer="100" to-port="4" />
<edge from-layer="100" from-port="5" to-layer="101" to-port="1" />
<edge from-layer="101" from-port="2" to-layer="102" to-port="0" />
<edge from-layer="102" from-port="2" to-layer="103" to-port="1" />
<edge from-layer="103" from-port="2" to-layer="175" to-port="1" />
<edge from-layer="104" from-port="0" to-layer="105" to-port="1" />
<edge from-layer="105" from-port="2" to-layer="107" to-port="0" />
<edge from-layer="106" from-port="0" to-layer="107" to-port="1" />
<edge from-layer="107" from-port="2" to-layer="109" to-port="0" />
<edge from-layer="108" from-port="0" to-layer="109" to-port="1" />
<edge from-layer="109" from-port="2" to-layer="175" to-port="2" />
<edge from-layer="110" from-port="0" to-layer="134" to-port="0" />
<edge from-layer="111" from-port="0" to-layer="112" to-port="1" />
<edge from-layer="112" from-port="2" to-layer="145" to-port="0" />
<edge from-layer="112" from-port="2" to-layer="133" to-port="0" />
<edge from-layer="113" from-port="0" to-layer="119" to-port="0" />
<edge from-layer="114" from-port="0" to-layer="115" to-port="1" />
<edge from-layer="115" from-port="2" to-layer="117" to-port="0" />
<edge from-layer="115" from-port="2" to-layer="172" to-port="3" />
<edge from-layer="115" from-port="2" to-layer="172" to-port="2" />
<edge from-layer="116" from-port="0" to-layer="117" to-port="1" />
<edge from-layer="117" from-port="2" to-layer="119" to-port="1" />
<edge from-layer="118" from-port="0" to-layer="119" to-port="2" />
<edge from-layer="119" from-port="3" to-layer="126" to-port="0" />
<edge from-layer="120" from-port="0" to-layer="122" to-port="0" />
<edge from-layer="121" from-port="0" to-layer="122" to-port="2" />
<edge from-layer="122" from-port="3" to-layer="125" to-port="0" />
<edge from-layer="123" from-port="0" to-layer="125" to-port="1" />
<edge from-layer="124" from-port="0" to-layer="125" to-port="2" />
<edge from-layer="125" from-port="3" to-layer="126" to-port="1" />
<edge from-layer="126" from-port="2" to-layer="128" to-port="0" />
<edge from-layer="127" from-port="0" to-layer="128" to-port="1" />
<edge from-layer="128" from-port="2" to-layer="130" to-port="0" />
<edge from-layer="129" from-port="0" to-layer="130" to-port="1" />
<edge from-layer="130" from-port="2" to-layer="132" to-port="0" />
<edge from-layer="131" from-port="0" to-layer="132" to-port="1" />
<edge from-layer="132" from-port="2" to-layer="133" to-port="1" />
<edge from-layer="133" from-port="2" to-layer="134" to-port="1" />
<edge from-layer="134" from-port="2" to-layer="168" to-port="0" />
<edge from-layer="135" from-port="1" to-layer="144" to-port="0" />
<edge from-layer="136" from-port="0" to-layer="138" to-port="1" />
<edge from-layer="137" from-port="0" to-layer="138" to-port="2" />
<edge from-layer="138" from-port="3" to-layer="139" to-port="0" />
<edge from-layer="139" from-port="1" to-layer="160" to-port="0" />
<edge from-layer="139" from-port="1" to-layer="141" to-port="0" />
<edge from-layer="140" from-port="0" to-layer="141" to-port="1" />
<edge from-layer="141" from-port="2" to-layer="143" to-port="0" />
<edge from-layer="142" from-port="0" to-layer="143" to-port="1" />
<edge from-layer="143" from-port="2" to-layer="144" to-port="1" />
<edge from-layer="144" from-port="2" to-layer="163" to-port="0" />
<edge from-layer="145" from-port="1" to-layer="162" to-port="0" />
<edge from-layer="146" from-port="0" to-layer="151" to-port="0" />
<edge from-layer="147" from-port="0" to-layer="149" to-port="1" />
<edge from-layer="148" from-port="0" to-layer="149" to-port="2" />
<edge from-layer="149" from-port="3" to-layer="151" to-port="1" />
<edge from-layer="149" from-port="3" to-layer="170" to-port="0" />
<edge from-layer="150" from-port="0" to-layer="151" to-port="2" />
<edge from-layer="151" from-port="3" to-layer="153" to-port="0" />
<edge from-layer="152" from-port="0" to-layer="153" to-port="1" />
<edge from-layer="153" from-port="2" to-layer="155" to-port="0" />
<edge from-layer="154" from-port="0" to-layer="155" to-port="1" />
<edge from-layer="155" from-port="2" to-layer="157" to-port="0" />
<edge from-layer="156" from-port="0" to-layer="157" to-port="1" />
<edge from-layer="157" from-port="2" to-layer="158" to-port="0" />
<edge from-layer="158" from-port="1" to-layer="161" to-port="0" />
<edge from-layer="159" from-port="0" to-layer="160" to-port="1" />
<edge from-layer="159" from-port="0" to-layer="163" to-port="2" />
<edge from-layer="160" from-port="3" to-layer="161" to-port="1" />
<edge from-layer="161" from-port="2" to-layer="162" to-port="1" />
<edge from-layer="162" from-port="2" to-layer="166" to-port="0" />
<edge from-layer="162" from-port="2" to-layer="163" to-port="1" />
<edge from-layer="163" from-port="3" to-layer="165" to-port="0" />
<edge from-layer="164" from-port="0" to-layer="165" to-port="1" />
<edge from-layer="165" from-port="2" to-layer="167" to-port="0" />
<edge from-layer="166" from-port="1" to-layer="167" to-port="1" />
<edge from-layer="167" from-port="2" to-layer="168" to-port="1" />
<edge from-layer="168" from-port="2" to-layer="173" to-port="0" />
<edge from-layer="169" from-port="0" to-layer="170" to-port="1" />
<edge from-layer="170" from-port="2" to-layer="172" to-port="0" />
<edge from-layer="171" from-port="0" to-layer="172" to-port="1" />
<edge from-layer="172" from-port="4" to-layer="173" to-port="1" />
<edge from-layer="173" from-port="2" to-layer="175" to-port="3" />
<edge from-layer="174" from-port="0" to-layer="175" to-port="4" />
<edge from-layer="175" from-port="5" to-layer="177" to-port="0" />
<edge from-layer="176" from-port="0" to-layer="177" to-port="1" />
<edge from-layer="177" from-port="2" to-layer="179" to-port="0" />
<edge from-layer="178" from-port="0" to-layer="179" to-port="1" />
<edge from-layer="179" from-port="2" to-layer="181" to-port="0" />
<edge from-layer="180" from-port="0" to-layer="181" to-port="1" />
<edge from-layer="181" from-port="2" to-layer="182" to-port="1" />
<edge from-layer="182" from-port="2" to-layer="201" to-port="0" />
<edge from-layer="182" from-port="2" to-layer="192" to-port="0" />
<edge from-layer="182" from-port="2" to-layer="185" to-port="0" />
<edge from-layer="183" from-port="0" to-layer="191" to-port="0" />
<edge from-layer="184" from-port="0" to-layer="185" to-port="1" />
<edge from-layer="185" from-port="2" to-layer="187" to-port="0" />
<edge from-layer="186" from-port="0" to-layer="187" to-port="1" />
<edge from-layer="187" from-port="2" to-layer="189" to-port="0" />
<edge from-layer="188" from-port="0" to-layer="189" to-port="1" />
<edge from-layer="189" from-port="2" to-layer="190" to-port="0" />
<edge from-layer="190" from-port="1" to-layer="191" to-port="1" />
<edge from-layer="191" from-port="2" to-layer="192" to-port="1" />
<edge from-layer="192" from-port="2" to-layer="194" to-port="0" />
<edge from-layer="192" from-port="2" to-layer="197" to-port="0" />
<edge from-layer="193" from-port="0" to-layer="194" to-port="1" />
<edge from-layer="194" from-port="2" to-layer="195" to-port="0" />
<edge from-layer="195" from-port="1" to-layer="198" to-port="0" />
<edge from-layer="196" from-port="0" to-layer="197" to-port="1" />
<edge from-layer="197" from-port="2" to-layer="198" to-port="1" />
<edge from-layer="198" from-port="2" to-layer="200" to-port="0" />
<edge from-layer="199" from-port="0" to-layer="200" to-port="1" />
<edge from-layer="200" from-port="2" to-layer="201" to-port="1" />
<edge from-layer="201" from-port="2" to-layer="202" to-port="0" />
</edges>
<rt_info>
<info name="OpenVINO Runtime" value="2026.5.0-23326-6ce8fccc044" />
<Runtime_version value="2026.5.0-23326-6ce8fccc044" />
<conversion_parameters>
<framework value="pytorch" />
<is_python_object value="True" />
</conversion_parameters>
<optimum>
<diffusers_version value="0.40.0" />
<optimum_intel_version value="2.3.0.dev0+af9c853a" />
<optimum_version value="2.3.0" />
<pytorch_version value="2.14.1" />
<transformers_version value="5.10.4" />
</optimum>
<runtime_options>
<ACTIVATIONS_SCALE_FACTOR value="8.0" />
</runtime_options>
</rt_info>
</net>