Upload 10 files

1_Pooling/config.json

@@ -0,0 +1,7 @@
+{
+  "word_embedding_dimension": 312,
+  "pooling_mode_cls_token": true,
+  "pooling_mode_mean_tokens": false,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false
+}

README.md

@@ -1,3 +1,76 @@
 ---
+language:
+- ru
+- en
+
+pipeline_tag: sentence-similarity
+
+tags:
+- russian
+- pretraining
+- embeddings
+- tiny
+- feature-extraction
+- sentence-similarity
+- retrieval
+- sentence-transformers
+- transformers
+- mteb
+
+datasets:
+- IlyaGusev/gazeta
+- zloelias/lenta-ru
+- HuggingFaceFW/fineweb-2
+- HuggingFaceFW/fineweb
+
 license: mit
+
+base_model: sergeyzh/rubert-tiny-turbo
+
 ---
+
+Быстрая модель BERT для задач текстового поиска (retrieval). Модель получена дистилляцией эмбеддингов русских и английских текстов [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3) в [rubert-tiny-turbo](https://huggingface.co/sergeyzh/rubert-tiny-turbo).
+
+Основные характеристики модели: 
+ - размер ембеддинга - 312, 
+ - длина контекста - 512, 
+ - слоёв - 3,
+ - префиксы - не требуются. 
+
+
+
+## Использование
+
+```Python
+from sentence_transformers import SentenceTransformer
+
+model = SentenceTransformer('sergeyzh/rubert-tiny-retriever')
+
+sentences = ["привет мир", "hello world", "здравствуй вселенная"]
+embeddings = model.encode(sentences)
+print(model.similarity(embeddings, embeddings))
+```
+
+
+
+## Метрики
+
+Оценки модели на задачах текстового поиска для русского языка:
+
+
+| Model Name | MIRACLReranking | MIRACLRetrival | RiaNewsRetrieval | RuBQReranking | RuBQRetrieval | Average |
+| :--- | :---: | :---: | :---: | :---: | :---: | :---: |
+| bge-m3 | 0,654 | 0,702 | 0,830 | 0,740 | 0,712 | 0,728 |
+| **rubert-tiny-retriever** | 0,574 | 0,530 | 0,611 | 0,668 | 0,589 | 0,594 |
+| rubert-tiny-turbo | 0,477 | 0,371 | 0,513 | 0,622 | 0,517 | 0,500 |
+| rubert-tiny2 | 0,158 | 0,019 | 0,140 | 0,461 | 0,109 | 0,177 |
+
+Оценки модели на задачах текстового поиска для английского языка:
+
+| Model Name | AILA Statutes | Argu Ana | Legal Bench Corporate Lobbying | SCIDOCS | Stack Overflow QA | Statcan Dialogue Dataset Retrieval | Wikipedia Retrieval Multilingual | Average |
+| :--- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
+| bge-m3 | 0,290 | 0,540 | 0,903 | 0,163 | 0,806 | 0,219 | 0,899 | 0,546 |
+| **rubert-tiny-retriever** | 0,161 | 0,432 | 0,862 | 0,094 | 0,454 | 0,103 | 0,880 | 0,426 |
+| rubert-tiny-turbo | 0,136 | 0,320 | 0,700 | 0,041 | 0,320 | 0,007 | 0,298 | 0,260 |
+| rubert-tiny2 | 0,138 | 0,277 | 0,602 | 0,012 | 0,200 | 0,004 | 0,145 | 0,197 |
+

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "sergeyzh/rubert-tiny-retriever",
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "emb_size": 312,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 312,
+  "initializer_range": 0.02,
+  "intermediate_size": 600,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 2048,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 3,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.57.6",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 83828
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47279aa703403542fef001a249f96a7325365e0f27c7e9a9def77a28dcc91755
+size 116781184

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": false
+}

special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,65 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "max_length": 512,
+  "model_max_length": 2048,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}