aLSTM(depth-2)+RecurrentPolicy (WT2)

Closed weights University of Manchester,Alan Turing Institute 32M parameters May 2018

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 Manchester,Alan Turing Institute
Organisation type
Academia,Government
Country
United Kingdom of Great Britain and Northern Ireland
Published
22 May 2018
Authors
Sebastian Flennerhag, Hujun Yin, John Keane, Mark Elliot

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
32M

32M (Table 3)

Training data
2,000,000 tokens

"Both models are trained for 10 000 steps with a batch size of 50 and a learning rate of 0.003." "Without tuning for WT2, both outperform previously published results in 150 epochs (table 3) and converge to new state of the art performance in 190 epochs"

Epochs
190

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
7.3 × 10¹⁶ FLOP

6 FLOP / token / parameter * 32000000 parameters * 2000000 tokens * 190 epochs = 7.296e+16 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
Open source

BSD-3 license: https://github.com/flennerhag/alstm/tree/master/examples

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
SOTA improvement

"Without tuning for WT2, both outperform previously published results in 150 epochs (table 3) and converge to new state of the art performance in 190 epochs"

Record confidence
Confident
Citations
12
Benchmark data
aLSTM(depth-2)+RecurrentPolicy (WT2)

Sources

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

Reference
Breaking the Activation Function Bottleneck through Adaptive Parameterization
Last updated
11 February 2026

What the numbers mean

Where it came from

aLSTM(depth-2)+RecurrentPolicy (WT2) was published by University of Manchester,Alan Turing Institute, in United Kingdom of Great Britain and Northern Ireland, in May 2018. It comes out of academia,Government.

It works in Language, and is recorded as doing language modeling.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

Training and provenance

The training run consumed about 7.3 × 10¹⁶ FLOP. That figure describes the cost of creating it and has no bearing on how quickly it generates text.

It was trained on about 2,000,000 tokens of text.

The reason it appears in this catalogue at all is sOTA improvement.

Answers

aLSTM(depth-2)+RecurrentPolicy (WT2) — common questions

01

What GPU do I need to run aLSTM(depth-2)+RecurrentPolicy (WT2)?

None. aLSTM(depth-2)+RecurrentPolicy (WT2) 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 aLSTM(depth-2)+RecurrentPolicy (WT2) open source?

No. aLSTM(depth-2)+RecurrentPolicy (WT2) has not had its weights published, so it exists only as a service controlled by its owner.

03

How many parameters does aLSTM(depth-2)+RecurrentPolicy (WT2) have?

aLSTM(depth-2)+RecurrentPolicy (WT2) has 32M parameters. 32M (Table 3). 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 aLSTM(depth-2)+RecurrentPolicy (WT2)?

aLSTM(depth-2)+RecurrentPolicy (WT2) was published by University of Manchester,Alan Turing Institute, based in United Kingdom of Great Britain and Northern Ireland, categorised as academia,Government.

05

When was aLSTM(depth-2)+RecurrentPolicy (WT2) released?

aLSTM(depth-2)+RecurrentPolicy (WT2) was published in May 2018. 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 aLSTM(depth-2)+RecurrentPolicy (WT2) used for?

aLSTM(depth-2)+RecurrentPolicy (WT2) works in Language, and is recorded as handling language modeling. These are the areas it was designed around; they describe intent rather than a hard boundary.

07

How much compute was used to train aLSTM(depth-2)+RecurrentPolicy (WT2)?

Around 7.3 × 10¹⁶ 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 11 February 2026

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.