B2T connection (16L)
No estimate
No hardware requirements for this model
This model's weights are open, but no parameter count has been published for it. Every memory and speed figure starts from that number, so we would rather show nothing than a fabricated estimate.
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
- LINE Corporation,Tohoku University
- Organisation type
- Industry,Academia
- Country
- Japan
- Published
- 1 June 2022
- Authors
- Sho Takase, Shun Kiyono, Sosuke Kobayashi, Jun Suzuki
What it does
The problem areas the model was built for. A model can carry several of each.
- Domain
- Language
- Task
- Language modeling/generation, Question answering
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.
- Training data
- 103,000,000 tokens
- Epochs
- 95
"We used WikiText-103 (Merity et al., 2017), which consists of a large number of tokens. The training, validation, and test sets contain 103M, 0.2M, and 0.2M tokens, respectively" Table 6 updates: 50K tokens/GPU: 1024 GPUs (Table 7): 192 50000*1024*192/103000000 = 95 epochs
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
- 2.8 × 10¹⁹ FLOP
- How it was established
- Hardware,Operation counting
192*7 GPU (P100) hours per Table 6 19 TFLOP/s 19 trillion * 192 * 7 * 3600 * 0.3 = 2.76e19 6 FLOP / parameter / token * 247000000 parameters [not sure how it was estimated] * 50000 steps * 1024 tokens / GPU * 192 GPUs = 1.4568653e+19 FLOP
The training run
What it physically took to train: which chips, how many, for how long, and what that drew from the wall.
- Training hardware
- NVIDIA P100
- Chips used
- 192
- Wall-clock time
- 7 hours
- Power draw
- 96.3 kW
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
- Open — downloadable
- Model access
- Open weights (unrestricted)
- Training code
- Open source
code and weights, MIT: https://github.com/takase/b2t_connection?tab=readme-ov-file
How it is classified
Labels the source dataset applies when tracking notable models, and how confident it is in the entry.
- Record confidence
- Confident
- Citations
- 18
- Benchmark data
- B2T connection (16L)
Sources
Where this record came from and when it was last checked.
- Reference
- On Layer Normalizations and Residual Connections in Transformers
- Last updated
- 25 May 2026
What the numbers mean
Background
B2T connection (16L) was published by LINE Corporation,Tohoku University, in the country recorded as Japan, during June 2022. The category the publisher falls under is industry,Academia.
It works in the domain of Language, and is recorded as performing the task of language modeling/generation, Question answering.
The weights are published, so it can be downloaded and run on your own hardware indefinitely, offline, with no account attached.
Training and provenance
The training run consumed about 2.8 × 10¹⁹ FLOP, on hardware recorded as NVIDIA P100. That figure measures what producing the model cost, and has no bearing on how fast it answers.
It was trained on a corpus of about 103,000,000 tokens of text.
Answers
B2T connection (16L) — common questions
B2T connection (16L)— where can I download it?
The weights are published, though we do not hold a repository link for it. This site calculates hardware requirements rather than hosting model files.
B2T connection (16L)— how much compute was used to train it?
Training consumed around 2.8 × 10¹⁹ FLOP, on hardware recorded as NVIDIA P100. 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.
B2T connection (16L)— what GPU do I need to run it?
We cannot say. It has open weights, but no parameter count has been published for it, and every memory and speed calculation starts from that number. We would rather show nothing than a fabricated estimate.
B2T connection (16L)— is it open source?
Its weights are published, so it can be downloaded and run on your own hardware. Note that open weights is not the same as open source in the full sense — it says nothing about the training data, the training code, or the commercial terms attached.
B2T connection (16L)— how many parameters does it have?
No parameter count has been published for it, which is why no memory or speed figure appears on this page.
B2T connection (16L)— who created it?
It was published by LINE Corporation,Tohoku University, based in Japan, an organisation categorised as industry,Academia.
B2T connection (16L)— when was it released?
It was published in June 2022. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.
B2T connection (16L)— what is it used for?
It works in the domain of Language, and is recorded as handling the task of language modeling/generation, Question answering. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.
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.