Upload folder using huggingface_hub

README.md

@@ -1,12 +1,11 @@
 ---
 license: gemma
 license_link: https://choosealicense.com/licenses/gemma/
-base_model:
-- google/gemma-2b-it
+base_model: google/gemma-2b-it
 base_model_relation: quantized
 ---
 # gemma-2b-it-int4-ov
-* Model creator: [Google](https://huggingface.co/google)
+* Model creator: [google](https://huggingface.co/google)
  * Original model: [gemma-2b-it](https://huggingface.co/google/gemma-2b-it)
 
 ## Description
@@ -27,10 +26,10 @@ For more information on quantization, check the [OpenVINO model optimization gui
 
 The provided OpenVINO™ IR model is compatible with:
 
-* OpenVINO version 2024.4.0 and higher
-* Optimum Intel 1.20.0 and higher
+* OpenVINO version 2024.5.0 and higher
+* Optimum Intel 1.21.0 and higher
 
-## Running Model Inference with [Optimum Intel](https://huggingface.co/docs/optimum/intel/index)
+## Running Model Inference
 
 1. Install packages required for using [Optimum Intel](https://huggingface.co/docs/optimum/intel/index) integration with the OpenVINO backend:
 
@@ -57,37 +56,6 @@ print(text)
 
 For more examples and possible optimizations, refer to the [OpenVINO Large Language Model Inference Guide](https://docs.openvino.ai/2024/learn-openvino/llm_inference_guide.html).
 
-## Running Model Inference with [OpenVINO GenAI](https://github.com/openvinotoolkit/openvino.genai)
-
-1. Install packages required for using OpenVINO GenAI.
-```
-pip install openvino-genai huggingface_hub
-```
-
-2. Download model from HuggingFace Hub
-   
-```
-import huggingface_hub as hf_hub
-
-model_id = "OpenVINO/gemma-2b-it-int4-ov"
-model_path = "gemma-2b-it-int4-ov"
-
-hf_hub.snapshot_download(model_id, local_dir=model_path)
-
-```
-
-3. Run model inference:
-
-```
-import openvino_genai as ov_genai
-
-device = "CPU"
-pipe = ov_genai.LLMPipeline(model_path, device)
-print(pipe.generate("What is OpenVINO?", max_length=200))
-```
-
-More GenAI usage examples can be found in OpenVINO GenAI library [docs](https://github.com/openvinotoolkit/openvino.genai/blob/master/src/README.md) and [samples](https://github.com/openvinotoolkit/openvino.genai?tab=readme-ov-file#openvino-genai-samples)
-
 ## Limitations
 
 Check the original model card for [original model card](https://huggingface.co/google/gemma-2b-it) for limitations.
@@ -98,4 +66,4 @@ The original model is distributed under [gemma](https://choosealicense.com/licen
 
 ## Disclaimer
 
-Intel is committed to respecting human rights and avoiding causing or contributing to adverse impacts on human rights. See [Intel’s Global Human Rights Principles](https://www.intel.com/content/dam/www/central-libraries/us/en/documents/policy-human-rights.pdf). Intel’s products and software are intended only to be used in applications that do not cause or contribute to adverse impacts on human rights.
+Intel is committed to respecting human rights and avoiding causing or contributing to adverse impacts on human rights. See [Intel’s Global Human Rights Principles](https://www.intel.com/content/dam/www/central-libraries/us/en/documents/policy-human-rights.pdf). Intel’s products and software are intended only to be used in applications that do not cause or contribute to adverse impacts on human rights.

config.json

@@ -1,5 +1,6 @@
 {
-  "_name_or_path": "google/gemma-2b-it",
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "OpenVINO/gemma-2b-it-int4-ov",
   "architectures": [
     "GemmaForCausalLM"
   ],
@@ -23,7 +24,7 @@
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.45.2",
+  "transformers_version": "4.46.3",
   "use_cache": true,
   "vocab_size": 256000
 }

generation_config.json

@@ -3,5 +3,5 @@
   "bos_token_id": 2,
   "eos_token_id": 1,
   "pad_token_id": 0,
-  "transformers_version": "4.45.2"
+  "transformers_version": "4.46.3"
 }

openvino_detokenizer.xml

@@ -1,16 +1,16 @@
 <?xml version="1.0"?>
 <net name="detokenizer" version="11">
 	<layers>
-		<layer id="0" name="Parameter_64251" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_48275" type="Parameter" version="opset1">
 			<data shape="?,?" element_type="i64" />
 			<output>
-				<port id="0" precision="I64" names="Parameter_64251">
+				<port id="0" precision="I64" names="Parameter_48275">
 					<dim>-1</dim>
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="1" name="Convert_64268" type="Convert" version="opset1">
+		<layer id="1" name="Convert_48292" type="Convert" version="opset1">
 			<data destination_type="i32" />
 			<input>
 				<port id="0" precision="I64">
@@ -25,7 +25,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="2" name="Constant_64218" type="Const" version="opset1">
+		<layer id="2" name="Constant_48242" type="Const" version="opset1">
 			<data element_type="u8" shape="2955910" offset="0" size="2955910" />
 			<output>
 				<port id="0" precision="U8">
@@ -33,7 +33,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="3" name="StringTensorUnpack_64219" type="StringTensorUnpack" version="extension">
+		<layer id="3" name="StringTensorUnpack_48243" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -52,7 +52,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="4" name="VocabDecoder_64252" type="VocabDecoder" version="extension">
+		<layer id="4" name="VocabDecoder_48276" type="VocabDecoder" version="extension">
 			<data skip_tokens="0, 1, 2, 3, 106, 107" />
 			<input>
 				<port id="0" precision="I32">
@@ -87,7 +87,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="5" name="Constant_64254" type="Const" version="opset1">
+		<layer id="5" name="Constant_48278" type="Const" version="opset1">
 			<data element_type="u8" shape="3" offset="2955910" size="3" />
 			<output>
 				<port id="0" precision="U8">
@@ -95,7 +95,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="6" name="Constant_64256" type="Const" version="opset1">
+		<layer id="6" name="Constant_48280" type="Const" version="opset1">
 			<data element_type="u8" shape="1" offset="2955913" size="1" />
 			<output>
 				<port id="0" precision="U8">
@@ -103,7 +103,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="7" name="RegexNormalization_64257" type="RegexNormalization" version="extension">
+		<layer id="7" name="RegexNormalization_48281" type="RegexNormalization" version="extension">
 			<data global_replace="true" />
 			<input>
 				<port id="0" precision="I32">
@@ -134,7 +134,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="8" name="ByteFallback_64258" type="ByteFallback" version="extension">
+		<layer id="8" name="ByteFallback_48282" type="ByteFallback" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -158,7 +158,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="9" name="FuzeRagged_64259" type="FuzeRagged" version="extension">
+		<layer id="9" name="FuzeRagged_48283" type="FuzeRagged" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -182,7 +182,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="10" name="StringTensorPack_64260" type="StringTensorPack" version="extension">
+		<layer id="10" name="StringTensorPack_48284" type="StringTensorPack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="I32">
@@ -201,7 +201,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="11" name="Result_64261" type="Result" version="opset1">
+		<layer id="11" name="Result_48285" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="STRING">
 					<dim>-1</dim>
@@ -234,10 +234,30 @@
 		<edge from-layer="10" from-port="3" to-layer="11" to-port="0" />
 	</edges>
 	<rt_info>
+		<add_attention_mask value="True" />
+		<add_prefix_space />
+		<add_special_tokens value="True" />
 		<bos_token_id value="2" />
 		<chat_template value="{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '&lt;start_of_turn>' + role + '&#10;' + message['content'] | trim + '&lt;end_of_turn>&#10;' }}{% endfor %}{% if add_generation_prompt %}{{'&lt;start_of_turn>model&#10;'}}{% endif %}" />
+		<clean_up_tokenization_spaces />
+		<detokenizer_input_type value="i64" />
 		<eos_token_id value="1" />
+		<handle_special_tokens_with_re />
+		<number_of_inputs value="1" />
+		<openvino_tokenizers_version value="2024.5.0.0" />
+		<openvino_version value="2024.5.0" />
 		<original_tokenizer_class value="&lt;class 'transformers.models.gemma.tokenization_gemma_fast.GemmaTokenizerFast'>" />
 		<pad_token_id value="0" />
+		<sentencepiece_version value="0.2.0" />
+		<skip_special_tokens value="True" />
+		<streaming_detokenizer value="False" />
+		<tiktoken_version value="0.8.0" />
+		<tokenizer_output_type value="i64" />
+		<tokenizers_version value="0.20.3" />
+		<transformers_version value="4.46.3" />
+		<use_max_padding value="False" />
+		<use_sentencepiece_backend value="False" />
+		<utf8_replace_mode />
+		<with_detokenizer value="True" />
 	</rt_info>
 </net>

openvino_tokenizer.xml

@@ -1,27 +1,27 @@
 <?xml version="1.0"?>
 <net name="tokenizer" version="11">
 	<layers>
-		<layer id="0" name="Parameter_64135" type="Parameter" version="opset1">
+		<layer id="0" name="Parameter_48159" type="Parameter" version="opset1">
 			<data shape="?" element_type="string" />
 			<output>
-				<port id="0" precision="STRING" names="Parameter_64135">
+				<port id="0" precision="STRING" names="Parameter_48159">
 					<dim>-1</dim>
 				</port>
 			</output>
 		</layer>
-		<layer id="1" name="Constant_64235" type="Const" version="opset1">
+		<layer id="1" name="Constant_48259" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="2" name="Constant_64236" type="Const" version="opset1">
+		<layer id="2" name="Constant_48260" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="4" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="3" name="Constant_64237" type="Const" version="opset1">
+		<layer id="3" name="Constant_48261" type="Const" version="opset1">
 			<data element_type="i32" shape="1" offset="8" size="4" />
 			<output>
 				<port id="0" precision="I32">
@@ -29,13 +29,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="4" name="Constant_64141" type="Const" version="opset1">
+		<layer id="4" name="Constant_48165" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="12" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="5" name="StringTensorUnpack_64136" type="StringTensorUnpack" version="extension">
+		<layer id="5" name="StringTensorUnpack_48160" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="STRING">
@@ -54,7 +54,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="6" name="ShapeOf_64137" type="ShapeOf" version="opset3">
+		<layer id="6" name="ShapeOf_48161" type="ShapeOf" version="opset3">
 			<data output_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -67,19 +67,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="7" name="Constant_64138" type="Const" version="opset1">
+		<layer id="7" name="Constant_48162" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="12" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="8" name="Constant_64139" type="Const" version="opset1">
+		<layer id="8" name="Constant_48163" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="12" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="9" name="Gather_64140" type="Gather" version="opset8">
+		<layer id="9" name="Gather_48164" type="Gather" version="opset8">
 			<data batch_dims="0" />
 			<input>
 				<port id="0" precision="I64">
@@ -92,13 +92,13 @@
 				<port id="3" precision="I64" />
 			</output>
 		</layer>
-		<layer id="10" name="Constant_64142" type="Const" version="opset1">
+		<layer id="10" name="Constant_48166" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="20" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="11" name="Range_64143" type="Range" version="opset4">
+		<layer id="11" name="Range_48167" type="Range" version="opset4">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -111,19 +111,19 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="12" name="Constant_64144" type="Const" version="opset1">
+		<layer id="12" name="Constant_48168" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="20" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="13" name="Constant_64145" type="Const" version="opset1">
+		<layer id="13" name="Constant_48169" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="20" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="14" name="Add_64146" type="Add" version="opset1">
+		<layer id="14" name="Add_48170" type="Add" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I64" />
@@ -133,13 +133,13 @@
 				<port id="2" precision="I64" />
 			</output>
 		</layer>
-		<layer id="15" name="Constant_64147" type="Const" version="opset1">
+		<layer id="15" name="Constant_48171" type="Const" version="opset1">
 			<data element_type="i64" shape="" offset="20" size="8" />
 			<output>
 				<port id="0" precision="I64" />
 			</output>
 		</layer>
-		<layer id="16" name="Range_64148" type="Range" version="opset4">
+		<layer id="16" name="Range_48172" type="Range" version="opset4">
 			<data output_type="i32" />
 			<input>
 				<port id="0" precision="I64" />
@@ -152,7 +152,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="17" name="Constant_64210" type="Const" version="opset1">
+		<layer id="17" name="Constant_48234" type="Const" version="opset1">
 			<data element_type="u8" shape="5282" offset="28" size="5282" />
 			<output>
 				<port id="0" precision="U8">
@@ -160,7 +160,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="18" name="SpecialTokensSplit_64211" type="SpecialTokensSplit" version="extension">
+		<layer id="18" name="SpecialTokensSplit_48235" type="SpecialTokensSplit" version="extension">
 			<input>
 				<port id="0" precision="I32">
 					<dim>-1</dim>
@@ -202,7 +202,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="19" name="Constant_64213" type="Const" version="opset1">
+		<layer id="19" name="Constant_48237" type="Const" version="opset1">
 			<data element_type="u8" shape="1" offset="5310" size="1" />
 			<output>
 				<port id="0" precision="U8">
@@ -210,7 +210,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="20" name="Constant_64215" type="Const" version="opset1">
+		<layer id="20" name="Constant_48239" type="Const" version="opset1">
 			<data element_type="u8" shape="3" offset="5311" size="3" />
 			<output>
 				<port id="0" precision="U8">
@@ -218,7 +218,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="21" name="RegexNormalization_64216" type="RegexNormalization" version="extension">
+		<layer id="21" name="RegexNormalization_48240" type="RegexNormalization" version="extension">
 			<data global_replace="true" />
 			<input>
 				<port id="0" precision="I32">
@@ -255,7 +255,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="22" name="Constant_64218" type="Const" version="opset1">
+		<layer id="22" name="Constant_48242" type="Const" version="opset1">
 			<data element_type="u8" shape="2955910" offset="5314" size="2955910" />
 			<output>
 				<port id="0" precision="U8">
@@ -263,7 +263,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="23" name="StringTensorUnpack_64219" type="StringTensorUnpack" version="extension">
+		<layer id="23" name="StringTensorUnpack_48243" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -282,7 +282,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="24" name="Constant_64224" type="Const" version="opset1">
+		<layer id="24" name="Constant_48248" type="Const" version="opset1">
 			<data element_type="u8" shape="5031736" offset="2961224" size="5031736" />
 			<output>
 				<port id="0" precision="U8">
@@ -290,7 +290,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="25" name="StringTensorUnpack_64225" type="StringTensorUnpack" version="extension">
+		<layer id="25" name="StringTensorUnpack_48249" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -309,7 +309,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="26" name="Constant_64227" type="Const" version="opset1">
+		<layer id="26" name="Constant_48251" type="Const" version="opset1">
 			<data element_type="u8" shape="4245743" offset="7992960" size="4245743" />
 			<output>
 				<port id="0" precision="U8">
@@ -317,7 +317,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="27" name="StringTensorUnpack_64228" type="StringTensorUnpack" version="extension">
+		<layer id="27" name="StringTensorUnpack_48252" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -336,7 +336,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="28" name="Constant_64221" type="Const" version="opset1">
+		<layer id="28" name="Constant_48245" type="Const" version="opset1">
 			<data element_type="u8" shape="4170" offset="12238703" size="4170" />
 			<output>
 				<port id="0" precision="U8">
@@ -344,7 +344,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="29" name="StringTensorUnpack_64222" type="StringTensorUnpack" version="extension">
+		<layer id="29" name="StringTensorUnpack_48246" type="StringTensorUnpack" version="extension">
 			<data mode="begins_ends" />
 			<input>
 				<port id="0" precision="U8">
@@ -363,7 +363,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="30" name="Constant_64229" type="Const" version="opset1">
+		<layer id="30" name="Constant_48253" type="Const" version="opset1">
 			<data element_type="i32" shape="216" offset="12242873" size="864" />
 			<output>
 				<port id="0" precision="I32">
@@ -371,7 +371,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="31" name="BPETokenizer_64230" type="BPETokenizer" version="extension">
+		<layer id="31" name="BPETokenizer_48254" type="BPETokenizer" version="extension">
 			<data unk_token="&lt;unk>" fuse_unk="true" suffix_indicator="" end_suffix="" byte_fallback="true" cache_capacity="51200" />
 			<input>
 				<port id="0" precision="I32">
@@ -441,7 +441,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="32" name="Subtract_64231" type="Subtract" version="opset1">
+		<layer id="32" name="Subtract_48255" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -457,13 +457,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="33" name="Constant_64232" type="Const" version="opset1">
+		<layer id="33" name="Constant_48256" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="12243737" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="34" name="Minimum_64233" type="Minimum" version="opset1">
+		<layer id="34" name="Minimum_48257" type="Minimum" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -477,7 +477,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="35" name="Subtract_64234" type="Subtract" version="opset1">
+		<layer id="35" name="Subtract_48258" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -493,7 +493,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="36" name="Constant_64238" type="Const" version="opset1">
+		<layer id="36" name="Constant_48262" type="Const" version="opset1">
 			<data element_type="i32" shape="2" offset="12" size="8" />
 			<output>
 				<port id="0" precision="I32">
@@ -501,7 +501,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="37" name="CombineSegments_64239" type="CombineSegments" version="extension">
+		<layer id="37" name="CombineSegments_48263" type="CombineSegments" version="extension">
 			<input>
 				<port id="0" precision="I32" />
 				<port id="1" precision="I32" />
@@ -542,7 +542,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="38" name="Subtract_64240" type="Subtract" version="opset1">
+		<layer id="38" name="Subtract_48264" type="Subtract" version="opset1">
 			<data auto_broadcast="numpy" />
 			<input>
 				<port id="0" precision="I32">
@@ -558,13 +558,13 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="39" name="Constant_64241" type="Const" version="opset1">
+		<layer id="39" name="Constant_48265" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="40" name="ReduceMax_64242" type="ReduceMax" version="opset1">
+		<layer id="40" name="ReduceMax_48266" type="ReduceMax" version="opset1">
 			<data keep_dims="false" />
 			<input>
 				<port id="0" precision="I32">
@@ -576,13 +576,13 @@
 				<port id="2" precision="I32" />
 			</output>
 		</layer>
-		<layer id="41" name="Constant_64243" type="Const" version="opset1">
+		<layer id="41" name="Constant_48267" type="Const" version="opset1">
 			<data element_type="i32" shape="" offset="0" size="4" />
 			<output>
 				<port id="0" precision="I32" />
 			</output>
 		</layer>
-		<layer id="42" name="RaggedToDense_64244" type="RaggedToDense" version="extension">
+		<layer id="42" name="RaggedToDense_48268" type="RaggedToDense" version="extension">
 			<data pad_right="false" />
 			<input>
 				<port id="0" precision="I32">
@@ -608,7 +608,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="43" name="Convert_64245" type="Convert" version="opset1">
+		<layer id="43" name="Convert_48269" type="Convert" version="opset1">
 			<data destination_type="i32" />
 			<input>
 				<port id="0" precision="BOOL">
@@ -623,7 +623,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="44" name="Convert_64245" type="Convert" version="opset1">
+		<layer id="44" name="Convert_48269" type="Convert" version="opset1">
 			<data destination_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -638,7 +638,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="46" name="RaggedToDense_64244.0" type="Convert" version="opset1">
+		<layer id="46" name="RaggedToDense_48268.0" type="Convert" version="opset1">
 			<data destination_type="i64" />
 			<input>
 				<port id="0" precision="I32">
@@ -653,7 +653,7 @@
 				</port>
 			</output>
 		</layer>
-		<layer id="47" name="Result_64248" type="Result" version="opset1">
+		<layer id="47" name="Result_48272" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="I64">
 					<dim>-1</dim>
@@ -661,7 +661,7 @@
 				</port>
 			</input>
 		</layer>
-		<layer id="45" name="Result_64250" type="Result" version="opset1">
+		<layer id="45" name="Result_48274" type="Result" version="opset1">
 			<input>
 				<port id="0" precision="I64">
 					<dim>-1</dim>
@@ -747,10 +747,30 @@
 		<edge from-layer="46" from-port="1" to-layer="47" to-port="0" />
 	</edges>
 	<rt_info>
+		<add_attention_mask value="True" />
+		<add_prefix_space />
+		<add_special_tokens value="True" />
 		<bos_token_id value="2" />
 		<chat_template value="{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '&lt;start_of_turn>' + role + '&#10;' + message['content'] | trim + '&lt;end_of_turn>&#10;' }}{% endfor %}{% if add_generation_prompt %}{{'&lt;start_of_turn>model&#10;'}}{% endif %}" />
+		<clean_up_tokenization_spaces />
+		<detokenizer_input_type value="i64" />
 		<eos_token_id value="1" />
+		<handle_special_tokens_with_re />
+		<number_of_inputs value="1" />
+		<openvino_tokenizers_version value="2024.5.0.0" />
+		<openvino_version value="2024.5.0" />
 		<original_tokenizer_class value="&lt;class 'transformers.models.gemma.tokenization_gemma_fast.GemmaTokenizerFast'>" />
 		<pad_token_id value="0" />
+		<sentencepiece_version value="0.2.0" />
+		<skip_special_tokens value="True" />
+		<streaming_detokenizer value="False" />
+		<tiktoken_version value="0.8.0" />
+		<tokenizer_output_type value="i64" />
+		<tokenizers_version value="0.20.3" />
+		<transformers_version value="4.46.3" />
+		<use_max_padding value="False" />
+		<use_sentencepiece_backend value="False" />
+		<utf8_replace_mode />
+		<with_detokenizer value="True" />
 	</rt_info>
 </net>