LBL

Closed weights University College London (UCL) 2M parameters June 2012

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 College London (UCL)
Organisation type
Academia
Country
United Kingdom of Great Britain and Northern Ireland
Published
27 June 2012
Authors
Andriy Mnih, Yee Whye Teh

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
2M

Vocabulary size (V) = 10,000 words Context size (c) = 2 words Feature dimension (d) = 100 Plugging this into the parameter count: Context word features: 10,000 x 100 = 1,000,000 Target word features: 10,000 x 100 = 1,000,000 Context weights: 2 x 100 x 100 = 20,000 Total parameters: 1,000,000 + 1,000,000 + 20,000 = 2,020,000 So for the LBL model configuration used in the experiments in this paper, with a 10K word vocabulary, 2-word context, and 100D feature vectors, the total number of parameter…

Training data
929,000 tokens

The paper does not explicitly state the number of epochs used for training the LBL models. However, it provides some relevant information: In Section 4, it mentions that the learning rates were adapted at the end of each epoch based on the validation set perplexity. In Section 4.1, it states: "NCE took about twice as many epochs as ML to converge." For maximum likelihood (ML) training, the training time is reported as 21 hours. So we can deduce: ML training took 21 hours and NCE took twice …

Epochs
45

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
5 × 10¹⁴ FLOP

6 FLOP / parameter / token * 2000000 parameters ['Likely' confidence] * 929000 tokens * 45 epochs ["Likely" confidence] = 5.0166e+14 FLOP

How it was established
Operation counting

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased
Training code
Unreleased

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Likely
Citations
835
Benchmark data
LBL

Sources

Where this record came from and when it was last checked.

Reference
A Fast and Simple Algorithm for Training Neural Probabilistic Language Models
Last updated
28 November 2025

What the numbers mean

About this model

LBL was published by University College London (UCL), in United Kingdom of Great Britain and Northern Ireland, in June 2012. academia is the category the publisher falls under.

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

The training run consumed about 5 × 10¹⁴ FLOP. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

The training set ran to roughly 929,000 tokens.

Answers

LBL — common questions

01

What GPU do I need to run LBL?

None. LBL 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.

02

Is LBL open source?

No. LBL has not had its weights published, so it exists only as a service controlled by its owner.

03

How many parameters does LBL have?

LBL has 2M parameters. Vocabulary size (V) = 10,000 words Context size (c) = 2 words Feature dimension (d) = 100 Plugging this into the parameter count: Context word features: 10,000 x 100 = 1,000,000 Target word features: 10,000 x 100 = 1,000,000 Context weights: 2 x 100 x 100 = 20,000 Total parameters: 1,000,000 + 1,000,000 + 20,000 = 2,020,000 So for the LBL model configuration used in the experiments in this paper, with a 10K word vocabulary, 2-word context, and 100D feature vectors, the total number of parameters is approximately 2 million. 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.

04

Who created LBL?

LBL was published by University College London (UCL), based in United Kingdom of Great Britain and Northern Ireland, categorised as academia.

05

When was LBL released?

LBL was published in June 2012. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

06

What is LBL used for?

LBL 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.

07

How much compute was used to train LBL?

Around 5 × 10¹⁴ FLOP. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

Source

Original publication

Record last updated 28 November 2025

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.