HLBL
No estimate
No hardware requirements for this model
The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.
On record
Full specification
Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.
Origin
Who built this model, where, and when it was published.
- Organisation
- University of Toronto
- Organisation type
- Academia
- Country
- Canada
- Published
- 8 December 2008
- Authors
- A. Mnih, Geoffrey E. Hinton
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling
Size
How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.
- Parameters
- 1.8M
- Training data
- 14,000,000 tokens
"Except for where stated otherwise, the models used for the experiments used 100 dimensional feature vectors and a context size of 5." "The vocabulary size for this dataset is 17964." Embedding: 17964 * 100 = 1796400 Context matrices: 5 * 100 * 100 = 50000 Unembedding: 0 (tied embedding “while the matrix of weights from the hidden layer to the output layer is simply the feature vector matrix R”) Total: 1796400 + 50000 = 1846400
"The dataset consists of a 14 million word training set"
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Why it is tracked
- Highly cited
- Record confidence
- Confident
Sources
Where this record came from and when it was last checked.
- Reference
- A Scalable Hierarchical Distributed Language Model
- Last updated
- 28 November 2025
What the numbers mean
What this model is
HLBL was published by University of Toronto, in Canada, in December 2008. The organisation is categorised as academia.
It works in Language, and is recorded as doing language modeling.
Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.
How it was trained
It was trained on about 14,000,000 tokens of text.
The reason it appears in this catalogue at all is highly cited.
Answers
HLBL — common questions
What is HLBL used for?
HLBL works in Language, and is recorded as handling language modeling. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.
What GPU do I need to run HLBL?
None. HLBL is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.
Is HLBL open source?
The licensing for HLBL was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.
How many parameters does HLBL have?
HLBL has 1.8M parameters. "Except for where stated otherwise, the models used for the experiments used 100 dimensional feature vectors and a context size of 5." "The vocabulary size for this dataset is 17964." Embedding: 17964 * 100 = 1796400 Context matrices: 5 * 100 * 100 = 50000 Unembedding: 0 (tied embedding “while the matrix of weights from the hidden layer to the output layer is simply the feature vector matrix R”) Total: 1796400 + 50000 = 1846400. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.
Who created HLBL?
HLBL was published by University of Toronto, based in Canada, categorised as academia.
When was HLBL released?
HLBL was published in December 2008. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
The other direction
Looking at it from the other side?
This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.