Musab6969/bge-small-vscode-dup

SentenceTransformer based on BAAI/bge-small-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, classification, clustering, and more.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: BAAI/bge-small-en-v1.5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'cls', 'include_prompt': True})
  (2): Normalize({})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    'User defined MCP servers are lost after June 2025 update\n\nAfter VSC insiders updated to the June 2025 update on its own, I do not see the MCP servers I added and I cannot add them again to use them.\nThe settings.json file is correct but the servers are not loaded anymore.\n\nPlease help.',
    '`chat.mcp.enabled` unexpectedly "managed by organization" after VS Code Insiders update\n\nThe `chat.mcp.enabled` setting in VS Code Insiders has become disabled and is now indicated as "managed by organization" following a recent update. This occurred despite the user not being connected to any organization through their Windows or GitHub accounts, and the setting was user-configurable before the update.\n\n![Image](https://github.com/user-attachments/assets/51e028b6-8ab1-4b42-8cf5-24a4ae3a78ae)\n\n**VS Code Insiders Version Details (Affected):**\n\n*   **Version:** 1.102.0-insider (user setup)\n*   **Commit:** 7e4e0f4e55d0d0a2a931aa4e0b7acc518e5da0dd\n*   **Date:** 2025-06-13T05:04:03.051Z\n*   **Electron:** 35.5.1\n*   **ElectronBuildId:** 11727614\n*   **Chromium:** 134.0.6998.205\n*   **Node.js:** 22.15.1\n*   **V8:** 13.4.114.21-electron.0\n*   **OS:** Windows_NT x64 10.0.26100\n\n**Previous VS Code Insiders Version (Unaffected):** 1.101.0.20250611\n\n**Current Stable VS Code Version (Unaffected):** 1.101.0\n\n**Steps to Reproduce:**\n\n1.  Have VS Code Insiders version 1.101.0.20250611 installed with `chat.mcp.enabled` either enabled or user-configurable.\n2.  Allow VS Code Insiders to auto-update to version 1.102.0.20250613.\n3.  Open VS Code Insiders settings.\n4.  Search for the setting `chat.mcp.enabled`.\n5.  (Optional) Check the same setting in VS Code stable version 1.101.0 to observe it remains user-configurable.\n\n**Expected Behavior:**\n\nThe `chat.mcp.enabled` setting should remain user-configurable in VS Code Insiders, allowing the user to enable or disable it as needed, consistent with its behavior in the stable release and previous Insiders versions.\n\n**Actual Behavior:**\n\nIn VS Code Insiders 1.102.0.20250613, the `chat.mcp.enabled` setting is disabled (off) and is now labeled as "managed by organization." The user is unable to change this setting. This change in behavior occurred directly after the VS Code Insiders update, without any changes to the user\'s system or account ',
    'Windows: custom titlebar context menu issues\n\nNeeds the revert of https://github.com/microsoft/vscode/pull/250631\n\nInitially reported by @benibenj, there are bugs around the custom title bar context menu on Windows:\n* often the first right-click does not show the menu and in that case the titlebar cannot be dragged anymore\n* sometimes the custom context menu appears in a wrong location',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.5474, 0.1768],
#         [0.5474, 1.0000, 0.1670],
#         [0.1768, 0.1670, 1.0000]])

Evaluation

Metrics

Semantic Similarity

• Dataset: heldout
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	nan
spearman_cosine	nan

Training Details

Training Dataset

Unnamed Dataset

• Size: 1,824 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 100 samples:

  	sentence_0	sentence_1
  type	string	string
  modality	text	text
  details	min: 7 tokens mean: 168.01 tokens max: 512 tokens	min: 7 tokens mean: 186.01 tokens max: 512 tokens

• Samples:

  sentence_0	sentence_1
  VS code crash Does this issue occur when all extensions are disabled?: Yes/No
  - VS Code Version: Version: 1.91.0 (1.91.0)
  - OS Version: OS Version: macOS 11.7.10 (20G1427)
  1. Just launching the vs code, crashes it and open a crash report popup
  Had tried to restart, clear cached data, tried universal as well as silicon version both ends up crashing
  here's a excerpt from crash report
  ```
  Process: Electron [1818]
  Path: /Applications/Visual Studio Code.app/Contents/MacOS/Electron
  Identifier: com.microsoft.VSCode
  Version: 1.91.0 (1.91.0)
  Code Type: ARM-64 (Native)
  Parent Process: ??? [1]
  Responsible: Electron [1818]
  User ID: 502
  Date/Time: 2024-07-06 23:37:28.080 +0530
  Report Version: 12
  Anonymous UUID: A507F1ED-34B4-79CC-640F-4825022BCC0C
  Time Awake Since Boot: 1200 seconds
  System Integrity Protectio...	v1.91 does not launch on Catalina or Big Sur due to Fatal error in V8 Does this issue occur when all extensions are disabled?: Yes/No
  - VS Code Version: 1.91
  - OS Version: Mac OS Catalina
  1. Launch 1.91 on Mac OS Catalina, VSCode will crash. If you try launching from the command line, you'll see the following error: "Fatal error in V8: v8::Template::Set Invalid value, must be a primitive or a Template"
  I know this is an old version of Mac OS, but if an OS will no longer be supported then perhaps it shouldn't update automatically. This particular installation is a developer test machine that won't/can't be updated past Catalina.
  Save Without Formatting still formats when saving with sudo Does this issue occur when all extensions are disabled?: Yes
  - VS Code Version: 1.87.2 (Commit: 863d2581ecda6849923a2118d93a088b0745d9d6)
  - OS Version: Linux Mint 21.3 Cinnamon (Linux x64 6.5.0-26-generic)
  1. Open VSCode as regular user without using sudo
  2. Make sure that Editor: Format on Save is enabled
  3. Edit a file that requires sudo to save
  4. Use Save Without Formatting function
  5. Click Retry as Sudo... when the option comes up
  6. Note that formatting is applied	"Retry as admin" don't respect "File: Save without Formatting" Open Visual Studio Code without admin privilege.
  In settings, check "Format On Save".
  Open a .html file that need admin privilege.
  Insert html code that need to be formatted.
  Save with "File: Save without Formatting"
  Visual Studio Code will show a message box with "Retry as admin..."
  Click on "Retry as admin..."
  Validate Windows popup warning about admin rights.
  The text will be formatted before be saved. (should not)
  The problem also appears with cmake-format (I didn't test more formatter).
  Extensions: none
  Remote Development Install Error Does this issue occur when all extensions are disabled?: Yes/No
  - VS Code Version:
  - OS Version:
  1. Update VSCode to 1.93 version
  2. Update Remote SSH extension to v0.114.1
  ```
  389523be2092: running
  Script executing under PID: 14946
  Installing to '/app/username'/.vscode-server...
  389523be2092%%1%%
  Downloading with wget
  wget is from busybox: no
  Program 'wget' appears to support flag '--no-config'
  Download complete
  389523be2092%%2%%
  tar --version: tar (GNU tar) 1.30
  Copyright (C) 2017 Free Software Foundation, Inc.
  License GPLv3+: GNU GPL version 3 or later https://gnu.org/licenses/gpl.html.
  This is free software: you are free to change and redistribute it.
  There is NO WARRANTY, to the extent permitted by law.
  Written by John Gilmore and Jay Fenlason.
  mv: cannot move 'code' to "'/app/username'/.vscode-server/code-4849ca9bdf9666755eb463db297b69e5385090e3": No such file or directory
  code 1.93.0 (commit 4849ca9...	ssh remote failed when "Remote.SSH: Server Install Path" configured Does this issue occur when all extensions are disabled?: Yes
  - VS Code Version:
  ```s
  Version: 1.93.0 (user setup)
  Commit: 4849ca9bdf9666755eb463db297b69e5385090e3
  Date: 2024-09-04T13:02:38.431Z
  Electron: 30.4.0
  ElectronBuildId: 10073054
  Chromium: 124.0.6367.243
  Node.js: 20.15.1
  V8: 12.4.254.20-electron.0
  OS: Windows_NT x64 10.0.22631
  ```
  - OS Version: local(window) , remote(centos 7.9)
  1. upgraded to my current version
  2. ssh remote connect to a Remote.SSH: Server Install Path configured host, my Remote.SSH: Server Install Path is
  3. vscode stuck at setup remote server, last message is some thing like 'scp server to remote'
  i reboot my remote server, cause vscode connection broken, and i saw messge
  ```s
  [10:30:08.370] Got error from ssh: spawn C:\WINDOWS\ssh.exe ENOENT
  [10:30:08.370] Chec...

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• per_device_train_batch_size: 32
• num_train_epochs: 3
• max_steps: -1
• learning_rate: 5e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0
• optim: adamw_torch_fused
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1
• label_smoothing_factor: 0.0
• bf16: False
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: False
• project: huggingface
• trackio_space_id: None
• trackio_bucket_id: None
• trackio_static_space_id: None
• per_device_eval_batch_size: 32
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: False
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: False
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• dataloader_multiprocessing_context: None
• dataloader_in_order: True
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_static_graph: None
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: None
• fsdp_config: None
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}
• warmup_ratio: None

Training Logs

Epoch	Step	heldout_spearman_cosine
1.0	57	nan
2.0	114	nan
3.0	171	nan

Training Time

• Training: 13.2 minutes

Framework Versions

• Python: 3.13.15
• Sentence Transformers: 5.7.0
• Transformers: 5.16.1
• PyTorch: 2.11.0+cu128
• Accelerate: 1.14.0
• Datasets: 4.8.5
• Tokenizers: 0.23.1

Additional Resources

• Training and Finetuning Embedding Models with Sentence Transformers: the end-to-end guide for training or finetuning Sentence Transformer models.
• Introduction to Matryoshka Embedding Models: variable-size embeddings that can be truncated with minimal quality loss.
• Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval: post-training compression of embedding vectors.
• Multimodal Embedding & Reranker Models with Sentence Transformers: use text, image, audio, and video models through the same API.
• Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers: train multimodal embedding models, with a Visual Document Retrieval walkthrough.

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MultipleNegativesRankingLoss

@misc{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}