Add new SentenceTransformer model

README.md

@@ -12,82 +12,75 @@ tags:
 - loss:MultipleNegativesRankingLoss
 base_model: nomic-ai/modernbert-embed-base
 widget:
-- source_sentence: What does 'multi-modal' refer to in the context of the services
-    mentioned?
+- source_sentence: What is the difference between traditional programming and ML?
   sentences:
-  - '1. Chatbots and Virtual Assistants
-
-    One of the most visible LLM integrations is in chatbots. Tools like ChatGPT, Claude,
-    and Bard are themselves chatbot interfaces built on LLMs. Many businesses are
-    now integrating these models into their websites and customer support systems.'
-  - For example, e-commerce websites can deploy LLM-powered assistants to help customers
-    find products, track orders, or get personalized recommendations—much more effectively
-    than traditional rule-based bots.
-  - Some services, like ColBERT, Marqo, and ColQwen, specialize in integrating LLMs
-    into search pipelines for both text and multi-modal (text + image) content.
-- source_sentence: What is one method mentioned for deploying LLMs?
-  sentences:
-  - However, deploying LLMs effectively in real-world applications often requires
-    LLM integration. This means embedding these models into systems, workflows, or
-    products where they can interact with other components like databases, APIs, user
-    interfaces, or even custom business logic
-  - Some services, like ColBERT, Marqo, and ColQwen, specialize in integrating LLMs
-    into search pipelines for both text and multi-modal (text + image) content.
-  - However, deploying LLMs effectively in real-world applications often requires
-    LLM integration. This means embedding these models into systems, workflows, or
-    products where they can interact with other components like databases, APIs, user
-    interfaces, or even custom business logic
-- source_sentence: What will an LLM likely respond with when prompted about the capital
-    of France?
-  sentences:
-  - . For instance, a spam filter doesn’t just block emails with specific keywords—it
-    learns from thousands of examples what spam typically looks like.
   - Over the past few years, the field of ML has advanced rapidly, especially in the
     area of Natural Language Processing (NLP)—the ability of machines to understand
     and generate human language. At the forefront of this progress are Large Language
     Models (LLMs), such as OpenAI’s GPT (Generative Pre-trained Transformer), Google’s
     PaLM, and Meta’s LLaMA
-  - For example, given a prompt like "The capital of France is", an LLM trained on
-    a wide range of texts will likely respond with "Paris". But beyond trivia, LLMs
-    can write essays, draft emails, simulate conversations, generate code snippets,
-    and much more.
-- source_sentence: What might an LLM be connected to in a customer support chatbot?
-  sentences:
-  - . For instance, a spam filter doesn’t just block emails with specific keywords—it
-    learns from thousands of examples what spam typically looks like.
   - . For example, integrating an LLM into a customer support chatbot might involve
     connecting it to a company’s internal knowledge base, enabling it to answer customer
     questions using accurate, up-to-date information.
-  - Large Language Models (LLMs) and Their Integrations
-- source_sentence: What type of dialogues can LLMs simulate?
+  - A major subset of AI is Machine Learning (ML), which involves algorithms that
+    learn from data rather than being explicitly programmed. Instead of writing detailed
+    instructions for every task, ML models find patterns in large datasets and use
+    these patterns to make predictions or decisions
+- source_sentence: What is one of the tasks mentioned that involves creating new written
+    content?
   sentences:
-  - 'Hallucinations: LLMs can sometimes generate plausible-sounding but incorrect
-    or fictional information.
-
-
-    Data Privacy: Sending sensitive data to third-party models raises privacy and
-    compliance concerns.
-
-
-    Cost and Latency: Running LLMs, especially large ones, can be computationally
-    expensive and slow.'
-  - '6. APIs and Developer Tools
-
-    Developers can integrate LLMs into their own apps using APIs provided by companies
-    like OpenAI, Anthropic, and Cohere. These APIs allow developers to send prompts
-    and receive intelligent outputs in return.
+  - In summary, AI and ML form the foundation for intelligent automation, while LLMs
+    represent a breakthrough in language understanding and generation. Integrating
+    these models into real-world systems unlocks practical value, turning raw intelligence
+    into tangible solutions
+  - '8. Security and Compliance Integrations
 
+    Some organizations are integrating LLMs to detect anomalies in text communications
+    (e.g., phishing detection or policy violations). LLMs can analyze language usage
+    and flag potentially suspicious behavior more flexibly than keyword-based filters.
 
-    This enables custom applications like:
 
+    Challenges in LLM Integration
 
-    Smart assistants in mobile apps
+    Despite their promise, integrating LLMs comes with challenges:'
+  - . These include text generation, summarization, translation, question answering,
+    code generation, and more.
+- source_sentence: What is one of the components mentioned alongside AI?
+  sentences:
+  - '2. Search Engines and Semantic Search
 
+    Traditional keyword-based search systems are being enhanced or replaced by semantic
+    search, where LLMs understand the meaning behind queries. Instead of just matching
+    words, they interpret intent.'
+  - For example, e-commerce websites can deploy LLM-powered assistants to help customers
+    find products, track orders, or get personalized recommendations—much more effectively
+    than traditional rule-based bots.
+  - Introduction to AI, Machine Learning, LLMs, and Their Integration
+- source_sentence: What is required to provide intelligent features within broader
+    applications?
+  sentences:
+  - . For instance, a spam filter doesn’t just block emails with specific keywords—it
+    learns from thousands of examples what spam typically looks like.
+  - 'The Rise of LLM Integrations
 
-    AI-powered research tools
+    While LLMs are powerful on their own, their true potential is unlocked through
+    integration—connecting these models with other software, services, or systems
+    to provide intelligent features within broader applications.
 
 
-    Voice interfaces'
+    Here are some key ways LLMs are being integrated into the digital world:'
+  - For instance, in a document management system, a user might type "policies about
+    sick leave", and the system—integrated with an LLM—could retrieve documents discussing
+    "medical leave", "employee absence", and "illness policies", even if those exact
+    words weren’t used.
+- source_sentence: What type of dialogues can LLMs simulate?
+  sentences:
+  - Companies are also experimenting with Retrieval-Augmented Generation (RAG)—a technique
+    where LLMs are paired with document databases (e.g., vector stores like Supabase,
+    Pinecone, or Weaviate) to answer questions with enterprise-specific knowledge.
+  - . For example, integrating an LLM into a customer support chatbot might involve
+    connecting it to a company’s internal knowledge base, enabling it to answer customer
+    questions using accurate, up-to-date information.
   - '5. Education and Learning Platforms
 
     Educational tools like Khanmigo (from Khan Academy) and other tutoring platforms
@@ -122,49 +115,49 @@ model-index:
       type: dim_768
     metrics:
     - type: cosine_accuracy@1
-      value: 0.6
+      value: 0.6666666666666666
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.8666666666666667
+      value: 0.8
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.8666666666666667
+      value: 1.0
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
       value: 1.0
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.6
+      value: 0.6666666666666666
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.28888888888888886
+      value: 0.2666666666666667
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.17333333333333337
+      value: 0.20000000000000007
       name: Cosine Precision@5
     - type: cosine_precision@10
       value: 0.10000000000000003
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.6
+      value: 0.6666666666666666
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.8666666666666667
+      value: 0.8
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.8666666666666667
+      value: 1.0
       name: Cosine Recall@5
     - type: cosine_recall@10
       value: 1.0
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.8025374182760189
+      value: 0.8310827786456928
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.74
+      value: 0.7766666666666667
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.74
+      value: 0.7766666666666667
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -180,10 +173,10 @@ model-index:
       value: 0.8
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.8
+      value: 0.8666666666666667
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.9333333333333333
+      value: 1.0
       name: Cosine Accuracy@10
     - type: cosine_precision@1
       value: 0.6666666666666666
@@ -192,10 +185,10 @@ model-index:
       value: 0.2666666666666667
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.16000000000000003
+      value: 0.17333333333333337
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.09333333333333335
+      value: 0.10000000000000003
       name: Cosine Precision@10
     - type: cosine_recall@1
       value: 0.6666666666666666
@@ -204,19 +197,19 @@ model-index:
       value: 0.8
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.8
+      value: 0.8666666666666667
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.9333333333333333
+      value: 1.0
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.7955687714024445
+      value: 0.8203966331432972
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.7527777777777779
+      value: 0.7651851851851852
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.7583333333333333
+      value: 0.7651851851851852
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -229,10 +222,10 @@ model-index:
       value: 0.6666666666666666
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.8
+      value: 0.8666666666666667
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.8
+      value: 1.0
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
       value: 1.0
@@ -241,10 +234,10 @@ model-index:
       value: 0.6666666666666666
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.2666666666666667
+      value: 0.28888888888888886
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.16000000000000003
+      value: 0.20000000000000007
       name: Cosine Precision@5
     - type: cosine_precision@10
       value: 0.10000000000000003
@@ -253,22 +246,22 @@ model-index:
       value: 0.6666666666666666
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.8
+      value: 0.8666666666666667
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.8
+      value: 1.0
       name: Cosine Recall@5
     - type: cosine_recall@10
       value: 1.0
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.7985736897839496
+      value: 0.8357043414408
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.7384126984126984
+      value: 0.7822222222222223
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.7384126984126984
+      value: 0.7822222222222223
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -278,49 +271,49 @@ model-index:
       type: dim_128
     metrics:
     - type: cosine_accuracy@1
-      value: 0.6666666666666666
+      value: 0.5333333333333333
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.8
+      value: 0.7333333333333333
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
       value: 0.8
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.8666666666666667
+      value: 0.9333333333333333
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.6666666666666666
+      value: 0.5333333333333333
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.2666666666666667
+      value: 0.2444444444444445
       name: Cosine Precision@3
     - type: cosine_precision@5
       value: 0.16000000000000003
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.08666666666666668
+      value: 0.09333333333333335
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.6666666666666666
+      value: 0.5333333333333333
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.8
+      value: 0.7333333333333333
       name: Cosine Recall@3
     - type: cosine_recall@5
       value: 0.8
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.8666666666666667
+      value: 0.9333333333333333
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.7700616222307202
+      value: 0.7203966331432973
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.74
+      value: 0.6540740740740741
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.7479365079365079
+      value: 0.6592022792022793
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -330,49 +323,49 @@ model-index:
       type: dim_64
     metrics:
     - type: cosine_accuracy@1
-      value: 0.6
+      value: 0.4666666666666667
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.8
+      value: 0.6666666666666666
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
       value: 0.8
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.8
+      value: 0.8666666666666667
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.6
+      value: 0.4666666666666667
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.2666666666666667
+      value: 0.22222222222222224
       name: Cosine Precision@3
     - type: cosine_precision@5
       value: 0.16000000000000003
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.08000000000000002
+      value: 0.08666666666666668
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.6
+      value: 0.4666666666666667
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.8
+      value: 0.6666666666666666
       name: Cosine Recall@3
     - type: cosine_recall@5
       value: 0.8
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.8
+      value: 0.8666666666666667
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.7174573004761944
+      value: 0.6507228370099043
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.6888888888888889
+      value: 0.5822222222222223
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.7003968253968255
+      value: 0.58890559732665
       name: Cosine Map@100
 ---
 
@@ -428,7 +421,7 @@ model = SentenceTransformer("Nuf-hugginface/modernbert-embed-quickb")
 sentences = [
     'What type of dialogues can LLMs simulate?',
     '5. Education and Learning Platforms\nEducational tools like Khanmigo (from Khan Academy) and other tutoring platforms are leveraging LLMs to provide real-time help to students. LLMs can break down complex topics, provide feedback on writing, and simulate Socratic-style dialogues.',
-    '6. APIs and Developer Tools\nDevelopers can integrate LLMs into their own apps using APIs provided by companies like OpenAI, Anthropic, and Cohere. These APIs allow developers to send prompts and receive intelligent outputs in return.\n\nThis enables custom applications like:\n\nSmart assistants in mobile apps\n\nAI-powered research tools\n\nVoice interfaces',
+    '. For example, integrating an LLM into a customer support chatbot might involve connecting it to a company’s internal knowledge base, enabling it to answer customer questions using accurate, up-to-date information.',
 ]
 embeddings = model.encode(sentences)
 print(embeddings.shape)
@@ -475,21 +468,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | dim_768    | dim_512    | dim_256    | dim_128    | dim_64     |
 |:--------------------|:-----------|:-----------|:-----------|:-----------|:-----------|
-| cosine_accuracy@1   | 0.6        | 0.6667     | 0.6667     | 0.6667     | 0.6        |
-| cosine_accuracy@3   | 0.8667     | 0.8        | 0.8        | 0.8        | 0.8        |
-| cosine_accuracy@5   | 0.8667     | 0.8        | 0.8        | 0.8        | 0.8        |
-| cosine_accuracy@10  | 1.0        | 0.9333     | 1.0        | 0.8667     | 0.8        |
-| cosine_precision@1  | 0.6        | 0.6667     | 0.6667     | 0.6667     | 0.6        |
-| cosine_precision@3  | 0.2889     | 0.2667     | 0.2667     | 0.2667     | 0.2667     |
-| cosine_precision@5  | 0.1733     | 0.16       | 0.16       | 0.16       | 0.16       |
-| cosine_precision@10 | 0.1        | 0.0933     | 0.1        | 0.0867     | 0.08       |
-| cosine_recall@1     | 0.6        | 0.6667     | 0.6667     | 0.6667     | 0.6        |
-| cosine_recall@3     | 0.8667     | 0.8        | 0.8        | 0.8        | 0.8        |
-| cosine_recall@5     | 0.8667     | 0.8        | 0.8        | 0.8        | 0.8        |
-| cosine_recall@10    | 1.0        | 0.9333     | 1.0        | 0.8667     | 0.8        |
-| **cosine_ndcg@10**  | **0.8025** | **0.7956** | **0.7986** | **0.7701** | **0.7175** |
-| cosine_mrr@10       | 0.74       | 0.7528     | 0.7384     | 0.74       | 0.6889     |
-| cosine_map@100      | 0.74       | 0.7583     | 0.7384     | 0.7479     | 0.7004     |
+| cosine_accuracy@1   | 0.6667     | 0.6667     | 0.6667     | 0.5333     | 0.4667     |
+| cosine_accuracy@3   | 0.8        | 0.8        | 0.8667     | 0.7333     | 0.6667     |
+| cosine_accuracy@5   | 1.0        | 0.8667     | 1.0        | 0.8        | 0.8        |
+| cosine_accuracy@10  | 1.0        | 1.0        | 1.0        | 0.9333     | 0.8667     |
+| cosine_precision@1  | 0.6667     | 0.6667     | 0.6667     | 0.5333     | 0.4667     |
+| cosine_precision@3  | 0.2667     | 0.2667     | 0.2889     | 0.2444     | 0.2222     |
+| cosine_precision@5  | 0.2        | 0.1733     | 0.2        | 0.16       | 0.16       |
+| cosine_precision@10 | 0.1        | 0.1        | 0.1        | 0.0933     | 0.0867     |
+| cosine_recall@1     | 0.6667     | 0.6667     | 0.6667     | 0.5333     | 0.4667     |
+| cosine_recall@3     | 0.8        | 0.8        | 0.8667     | 0.7333     | 0.6667     |
+| cosine_recall@5     | 1.0        | 0.8667     | 1.0        | 0.8        | 0.8        |
+| cosine_recall@10    | 1.0        | 1.0        | 1.0        | 0.9333     | 0.8667     |
+| **cosine_ndcg@10**  | **0.8311** | **0.8204** | **0.8357** | **0.7204** | **0.6507** |
+| cosine_mrr@10       | 0.7767     | 0.7652     | 0.7822     | 0.6541     | 0.5822     |
+| cosine_map@100      | 0.7767     | 0.7652     | 0.7822     | 0.6592     | 0.5889     |
 
 <!--
 ## Bias, Risks and Limitations
@@ -512,16 +505,16 @@ You can finetune this model on your own dataset.
 * Size: 127 training samples
 * Columns: <code>anchor</code> and <code>positive</code>
 * Approximate statistics based on the first 127 samples:
-  |         | anchor                                                                           | positive                                                                           |
-  |:--------|:---------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
-  | type    | string                                                                           | string                                                                             |
-  | details | <ul><li>min: 8 tokens</li><li>mean: 13.2 tokens</li><li>max: 20 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 53.85 tokens</li><li>max: 86 tokens</li></ul> |
+  |         | anchor                                                                            | positive                                                                           |
+  |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
+  | type    | string                                                                            | string                                                                             |
+  | details | <ul><li>min: 8 tokens</li><li>mean: 13.28 tokens</li><li>max: 25 tokens</li></ul> | <ul><li>min: 13 tokens</li><li>mean: 53.34 tokens</li><li>max: 86 tokens</li></ul> |
 * Samples:
-  | anchor                                                                           | positive                                                                                                                                                                                                                                                                                             |
-  |:---------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-  | <code>What documents could the system retrieve in relation to sick leave?</code> | <code>For instance, in a document management system, a user might type "policies about sick leave", and the system—integrated with an LLM—could retrieve documents discussing "medical leave", "employee absence", and "illness policies", even if those exact words weren’t used.</code>            |
-  | <code>What is one of the most visible integrations of LLM technology?</code>     | <code>1. Chatbots and Virtual Assistants<br>One of the most visible LLM integrations is in chatbots. Tools like ChatGPT, Claude, and Bard are themselves chatbot interfaces built on LLMs. Many businesses are now integrating these models into their websites and customer support systems.</code> |
-  | <code>What does AI stand for?</code>                                             | <code>Introduction to AI, Machine Learning, LLMs, and Their Integration</code>                                                                                                                                                                                                                       |
+  | anchor                                                                            | positive                                                                                                                                                                                                                                                                                                      |
+  |:----------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+  | <code>What task mentioned is related to providing answers to inquiries?</code>    | <code>. These include text generation, summarization, translation, question answering, code generation, and more.</code>                                                                                                                                                                                      |
+  | <code>What do LLMs learn to work effectively?</code>                              | <code>LLMs work by learning statistical relationships between words and phrases, allowing them to predict and generate language that feels natural. The power of these models lies not only in their size but also in the diversity of tasks they can perform with little to no task-specific training</code> |
+  | <code>In which industries is the generalization ability considered useful?</code> | <code>. This generalization ability makes them incredibly useful across industries—from customer service and education to software development and healthcare.</code>                                                                                                                                         |
 * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
   ```json
   {
@@ -680,13 +673,13 @@ You can finetune this model on your own dataset.
 </details>
 
 ### Training Logs
-| Epoch   | Step  | Training Loss | dim_768_cosine_ndcg@10 | dim_512_cosine_ndcg@10 | dim_256_cosine_ndcg@10 | dim_128_cosine_ndcg@10 | dim_64_cosine_ndcg@10 |
-|:-------:|:-----:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|
-| 1.0     | 4     | -             | 0.7853                 | 0.8214                 | 0.7673                 | 0.7586                 | 0.6883                |
-| **2.0** | **8** | **-**         | **0.7764**             | **0.7902**             | **0.7686**             | **0.7701**             | **0.7321**            |
-| 2.5     | 10    | 13.8004       | -                      | -                      | -                      | -                      | -                     |
-| 3.0     | 12    | -             | 0.8028                 | 0.7710                 | 0.7932                 | 0.7701                 | 0.7175                |
-| 4.0     | 16    | -             | 0.8025                 | 0.7956                 | 0.7986                 | 0.7701                 | 0.7175                |
+| Epoch   | Step   | Training Loss | dim_768_cosine_ndcg@10 | dim_512_cosine_ndcg@10 | dim_256_cosine_ndcg@10 | dim_128_cosine_ndcg@10 | dim_64_cosine_ndcg@10 |
+|:-------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|
+| 1.0     | 4      | -             | 0.7790                 | 0.7120                 | 0.7474                 | 0.6321                 | 0.5684                |
+| 2.0     | 8      | -             | 0.8275                 | 0.7966                 | 0.8091                 | 0.6904                 | 0.6102                |
+| 2.5     | 10     | 13.4453       | -                      | -                      | -                      | -                      | -                     |
+| 3.0     | 12     | -             | 0.8311                 | 0.8204                 | 0.8357                 | 0.7178                 | 0.6557                |
+| **4.0** | **16** | **-**         | **0.8311**             | **0.8204**             | **0.8357**             | **0.7204**             | **0.6507**            |
 
 * The bold row denotes the saved checkpoint.
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7fa83288c96da5c91a8d0a7f680fa88e3010ec41496910019839fb29c898aa3
+oid sha256:b3d2c21c46d543a74109f26cb657b70ff5920256bc2c79d47d87b9086e1fae84
 size 596070136