DistilBERT
E435865
DistilBERT is a smaller, faster, and lighter-weight distilled version of the BERT language model designed to retain most of its performance while being more efficient for practical NLP applications.
All labels observed (4)
| Label | Occurrences |
|---|---|
| DistilBERT canonical | 3 |
| A Lite BERT | 2 |
| DistilBERT: a distilled version of BERT: smaller, faster, cheaper and lighter | 1 |
| DistilBertModel | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T4389190 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: DistilBERT Context triple: [Hugging Face Transformers, supportsModelType, DistilBERT]
-
A.
Hugging Face Transformers
Hugging Face Transformers is a widely used open-source library that provides state-of-the-art transformer-based models and tools for natural language processing and related machine learning tasks.
-
B.
GPT-2
GPT-2 is a large transformer-based language model known for generating coherent, human-like text and sparking widespread discussion about the implications of advanced AI text generation.
-
C.
AllenNLP
AllenNLP is an open-source natural language processing research library built on PyTorch, designed to facilitate the development and evaluation of state-of-the-art NLP models.
-
D.
GPT-3
GPT-3 is a large-scale autoregressive language model known for generating human-like text and performing a wide range of natural language tasks with minimal fine-tuning.
-
E.
LLM
LLM is the ICAO airline designator assigned to Yamal Airlines, a Russian regional carrier.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: DistilBERT Target entity description: DistilBERT is a smaller, faster, and lighter-weight distilled version of the BERT language model designed to retain most of its performance while being more efficient for practical NLP applications.
-
A.
Hugging Face Transformers
Hugging Face Transformers is a widely used open-source library that provides state-of-the-art transformer-based models and tools for natural language processing and related machine learning tasks.
-
B.
GPT-2
GPT-2 is a large transformer-based language model known for generating coherent, human-like text and sparking widespread discussion about the implications of advanced AI text generation.
-
C.
AllenNLP
AllenNLP is an open-source natural language processing research library built on PyTorch, designed to facilitate the development and evaluation of state-of-the-art NLP models.
-
D.
GPT-3
GPT-3 is a large-scale autoregressive language model known for generating human-like text and performing a wide range of natural language tasks with minimal fine-tuning.
-
E.
LLM
LLM is the ICAO airline designator assigned to Yamal Airlines, a Russian regional carrier.
- F. None of above. chosen
Statements (49)
| Predicate | Object |
|---|---|
| instanceOf |
distilled model
ⓘ
neural network model ⓘ pretrained language model ⓘ transformer-based language model ⓘ |
| architectureType | Transformer ⓘ |
| availableInLibrary | Transformers ⓘ |
| basedOn | BERT ⓘ |
| compatibleWith |
Hugging Face Tokenizers
ⓘ
linked to:
Hugging Face Transformers
ONNX export ⓘ |
| designedFor |
efficiency
ⓘ
faster inference ⓘ lower memory usage ⓘ |
| developedBy | Hugging Face ⓘ |
| distilledFrom |
BERT base uncased
ⓘ
linked to:
BERT
|
| hasModelVariant |
distilbert-base-cased
ⓘ
distilbert-base-multilingual-cased ⓘ
linked to:
distilbert-base-cased
distilbert-base-uncased ⓘ distilbert-base-uncased-finetuned-sst-2-english ⓘ |
| hiddenSize | 768 ⓘ |
| implementedIn | PyTorch ⓘ |
| inputType | tokenized text ⓘ |
| language | English ⓘ |
| license | Apache-2.0 ⓘ |
| maintainedBy | Hugging Face ⓘ |
| numberOfAttentionHeads | 12 ⓘ |
| numberOfLayers | 6 ⓘ |
| outputType |
contextualized token embeddings
ⓘ
sequence representation ⓘ |
| paperArchive | arXiv ⓘ |
| paperArxivId | 1910.01108 ⓘ |
| paperTitle |
DistilBERT: a distilled version of BERT: smaller, faster, cheaper and lighter
ⓘ
linked to:
DistilBERT
|
| parameterCountRelativeTo | about 40 percent fewer parameters than BERT base ⓘ |
| performanceRetention | retains about 97 percent of BERT base performance on GLUE ⓘ |
| releasedBy | Hugging Face ⓘ |
| releaseYear | 2019 ⓘ |
| speedRelativeTo | about 60 percent faster than BERT base ⓘ |
| supportsTask |
feature extraction
ⓘ
named entity recognition ⓘ question answering ⓘ sentiment analysis ⓘ sequence labeling ⓘ text classification ⓘ |
| tokenizerType | WordPiece ⓘ |
| trainingObjective |
distillation from BERT
ⓘ
masked language modeling ⓘ |
| usedFor |
production NLP systems
ⓘ
resource-constrained environments ⓘ |
| usesMechanism | self-attention ⓘ |
| vocabularySize | 30522 ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: DistilBERT Description of subject: DistilBERT is a smaller, faster, and lighter-weight distilled version of the BERT language model designed to retain most of its performance while being more efficient for practical NLP applications.
Referenced by (7)
Full triples — surface form annotated when it differs from this entity's canonical label.
DistilBERT
→
paperTitle
→
DistilBERT: a distilled version of BERT: smaller, faster, cheaper and lighter
ⓘ
linked to: DistilBERT
linked to: DistilBERT
linked to: DistilBERT