LSTM + dynamic eval
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 Edinburgh
- Organisation type
- Academia
- Country
- United Kingdom of Great Britain and Northern Ireland
- Published
- 21 September 2017
- Authors
- Ben Krause, Emmanuel Kahembwe, Iain Murray, Steve Renals
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
- 50M
- Training data
- 90,000,000 tokens
table 2
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-2: https://github.com/benkrause/dynamic-evaluation
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
- Record confidence
- Likely
- Citations
- 147
- Benchmark data
- LSTM + dynamic eval
"Dynamic evaluation outperforms existing adaptation approaches in our comparisons. Dynamic evaluation improves the state-of-the-art word-level perplexities on the Penn Treebank and WikiText-2 datasets to 51.1 and 44.3 respectively"
Sources
Where this record came from and when it was last checked.
- Reference
- Dynamic Evaluation of Neural Sequence Models
- Last updated
- 25 May 2026
What the numbers mean
Background
LSTM + dynamic eval was published by University of Edinburgh, in United Kingdom of Great Britain and Northern Ireland, in September 2017. It comes out of academia.
It works in Language, and is recorded as doing language modeling.
This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.
What went into building it
Around 90,000,000 tokens went into training it.
It is tracked in the underlying dataset for one reason in particular: sOTA improvement.
Answers
LSTM + dynamic eval — common questions
Is LSTM + dynamic eval open source?
No. LSTM + dynamic eval has not had its weights published, so it exists only as a service controlled by its owner.
How many parameters does LSTM + dynamic eval have?
LSTM + dynamic eval has 50M parameters. table 2. 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 LSTM + dynamic eval?
LSTM + dynamic eval was published by University of Edinburgh, based in United Kingdom of Great Britain and Northern Ireland, categorised as academia.
When was LSTM + dynamic eval released?
LSTM + dynamic eval was published in September 2017. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
What is LSTM + dynamic eval used for?
LSTM + dynamic eval 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.
What GPU do I need to run LSTM + dynamic eval?
None. LSTM + dynamic eval 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.
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.