Towards Best Practices for Open Datasets for LLM Training
Paper
•
2501.08365
•
Published
•
51
BigScience Workshop
non-profit
AI & ML interests
A one-year long research workshop on large language models: the Summer of Language Models 21 🌸
Recent Activity
View all activity
bigscience's activity
Adding `safetensors` variant of this model
#2 opened 10 days ago
by
SFconvertbot
Adding `safetensors` variant of this model
1
#3 opened 10 days ago
by
SFconvertbot
lvwerra
authored
a
paper
15 days ago
salomey
authored
a
paper
15 days ago
Post
2963
💫...And we're live!💫 Seasonal newsletter from ethicsy folks at Hugging Face, exploring the ethics of "AI Agents"
https://huggingface.co/blog/ethics-soc-7
Our analyses found:
- There's a spectrum of "agent"-ness
- *Safety* is a key issue, leading to many other value-based concerns
Read for details & what to do next!
With @evijit , @giadap , and @sasha
https://huggingface.co/blog/ethics-soc-7
Our analyses found:
- There's a spectrum of "agent"-ness
- *Safety* is a key issue, leading to many other value-based concerns
Read for details & what to do next!
With @evijit , @giadap , and @sasha
Skylion007
authored
a
paper
21 days ago
dirkgr
authored
a
paper
25 days ago
Jekaterina
authored
2
papers
about 1 month ago
Adding `safetensors` variant of this model
#2 opened about 1 month ago
by
SFconvertbot
Adding `safetensors` variant of this model
#4 opened about 1 month ago
by
SFconvertbot
tomsherborne
authored
a
paper
about 1 month ago
Post
1219
Performance leap: TGI v3 is out. Processes 3x more tokens, 13x faster than vLLM on long prompts. Zero config !
3x more tokens.
By reducing our memory footprint, we’re able to ingest many more tokens and more dynamically than before. A single L4 (24GB) can handle 30k tokens on llama 3.1-8B, while vLLM gets barely 10k. A lot of work went into reducing the footprint of the runtime and its effect are best seen on smaller constrained environments.
13x faster
On long prompts (200k+ tokens) conversation replies take 27.5s in vLLM, while it takes only 2s in TGI. How so ? We keep the initial conversation around, so when a new reply comes in, we can answer almost instantly. The overhead of the lookup is ~5us. Thanks @Dani ël de Kok for the beast data structure.
Zero config
That’s it. Remove all the flags your are using and you’re likely to get the best performance. By evaluating the hardware and model, TGI carefully selects automatic values to give best performance. In production, we don’t have any flags anymore in our deployments. We kept all existing flags around, they may come in handy in niche scenarios.
Read more: https://huggingface.co/docs/text-generation-inference/conceptual/chunking
3x more tokens.
By reducing our memory footprint, we’re able to ingest many more tokens and more dynamically than before. A single L4 (24GB) can handle 30k tokens on llama 3.1-8B, while vLLM gets barely 10k. A lot of work went into reducing the footprint of the runtime and its effect are best seen on smaller constrained environments.
13x faster
On long prompts (200k+ tokens) conversation replies take 27.5s in vLLM, while it takes only 2s in TGI. How so ? We keep the initial conversation around, so when a new reply comes in, we can answer almost instantly. The overhead of the lookup is ~5us. Thanks @Dani ël de Kok for the beast data structure.
Zero config
That’s it. Remove all the flags your are using and you’re likely to get the best performance. By evaluating the hardware and model, TGI carefully selects automatic values to give best performance. In production, we don’t have any flags anymore in our deployments. We kept all existing flags around, they may come in handy in niche scenarios.
Read more: https://huggingface.co/docs/text-generation-inference/conceptual/chunking
christopher
posted
an
update
about 2 months ago
Post
1639
The folks at Foursquare released a dataset of 104.5 million places of interest (
foursquare/fsq-os-places) and here's all of them on a plot
-
- 4 replies
christopher
posted
an
update
about 2 months ago
Post
2384
The Lichess database of games, puzzles, and engine evaluations is now on the Hub: https://huggingface.co/Lichess
Billions of chess data points to download, query, and stream and we're excited to see what you'll build with it! ♟️ 🤗
- Lichess/positions-datasets-66f50837db5cd3287d60d489
- Lichess/games-datasets-66f508df78f4b43e1bb2d353
Billions of chess data points to download, query, and stream and we're excited to see what you'll build with it! ♟️ 🤗
- Lichess/positions-datasets-66f50837db5cd3287d60d489
- Lichess/games-datasets-66f508df78f4b43e1bb2d353
lvwerra
authored
a
paper
3 months ago
Adding `safetensors` variant of this model
#4 opened 3 months ago
by
SFconvertbot
Request: DOI
3
#43 opened 3 months ago
by
ovv4thewin
ruochenz
authored
a
paper
4 months ago
rbawden
authored
a
paper
4 months ago