LLMs Lost in Translation: M-ALERT uncovers Cross-Linguistic Safety Gaps
Paper
•
2412.15035
•
Published
•
4
Occiglot
community
AI & ML interests
Open Source Language Models for Europe
Recent Activity
View all activity
occiglot's activity
mbrack
authored
a
paper
2 days ago
PSaiml
authored
a
paper
2 days ago
barthfab
updated
a
dataset
6 days ago
malteos
updated
a
collection
16 days ago
Post
1127
My latest project is the outcome of the last 2+ years working with TPUs from the amazing TPU Research Cloud (TRC) program and training Encoder-only LMs with the TensorFlow Model Garden library.
👉 Link: https://github.com/stefan-it/model-garden-lms
An overview of some features:
- Cheatsheet for setting-up a TPU VM Pod (with all necessary dependencies) to pretrain LMs with TF Model Garden
- Conversion scripts that convert TF Model Garden weights to Hugging Face Transformers-compatible models
- Supported architectures include BERT, BERT with Token Dropping and TEAMS
I also released BERT-based models pretrained on the great Hugging Face FineWeb and FineWeb-Edu datasets (10BT subset). With more to come!
👉 Model Hub Link: https://huggingface.co/model-garden-lms
If you find these resources useful, please give them a like!
Made from Bavarian Oberland with ❤️ and 🥨.
👉 Link: https://github.com/stefan-it/model-garden-lms
An overview of some features:
- Cheatsheet for setting-up a TPU VM Pod (with all necessary dependencies) to pretrain LMs with TF Model Garden
- Conversion scripts that convert TF Model Garden weights to Hugging Face Transformers-compatible models
- Supported architectures include BERT, BERT with Token Dropping and TEAMS
I also released BERT-based models pretrained on the great Hugging Face FineWeb and FineWeb-Edu datasets (10BT subset). With more to come!
👉 Model Hub Link: https://huggingface.co/model-garden-lms
If you find these resources useful, please give them a like!
Made from Bavarian Oberland with ❤️ and 🥨.
mbrack
updated
a
dataset
19 days ago
BramVanroy
posted
an
update
19 days ago
Post
440
In the spirit of "Better late than never", I've finally written a brief overview paper for GEITje 7B Ultra. Initially released 10 months ago (oops), but still reaching around 1300 monthly downloads across the HF ecosystem (not including ollama).
GEITje 7B Ultra: A Conversational Model for Dutch (2412.04092)
While the paper discusses the model a little bit, I especially wanted to write about the datasets, which to this day seem an important asset for Dutch LLM training (SFT and preference tuning). We have a long way to go for Dutch, but publishing transparent and reproducible artefacts seems an important step to me, alongside having open discussions about data, bias, architectures.
In that spirit, thanks are in order for the creation of GEITje 7B Ultra and all related datasets:
- Michiel Buisman and UWV for providing the means to create the datasets
- Flemish Supercomputer Center (VSC) for the compute
- The Hugging Face Fellows and rest of the team for their discussions and insights
- The Dutch NLP community, notably @Rijgersberg for building the base GEITje model and the fruitful discussions we've had
More to come, step by step!
BramVanroy/geitje-7b-ultra-65c1ee010ad80fd1f6a8f208
GEITje 7B Ultra: A Conversational Model for Dutch (2412.04092)
While the paper discusses the model a little bit, I especially wanted to write about the datasets, which to this day seem an important asset for Dutch LLM training (SFT and preference tuning). We have a long way to go for Dutch, but publishing transparent and reproducible artefacts seems an important step to me, alongside having open discussions about data, bias, architectures.
In that spirit, thanks are in order for the creation of GEITje 7B Ultra and all related datasets:
- Michiel Buisman and UWV for providing the means to create the datasets
- Flemish Supercomputer Center (VSC) for the compute
- The Hugging Face Fellows and rest of the team for their discussions and insights
- The Dutch NLP community, notably @Rijgersberg for building the base GEITje model and the fruitful discussions we've had
More to come, step by step!
BramVanroy/geitje-7b-ultra-65c1ee010ad80fd1f6a8f208
mbrack
updated
a
dataset
22 days ago
mbrack
authored
a
paper
6 months ago
pjox
authored
a
paper
6 months ago
BramVanroy
posted
an
update
7 months ago
Post
1693
The InstructGPT paper mentions that they insert 10% pretraining data during SFT, which they find improves the effect of PPO (IIUC). Has anyone else done later ablations on this? I've only seen the inverse suggested, mixing in SFT data during pretraining.
- 2 replies
BramVanroy
posted
an
update
8 months ago
Post
2297
All my models seem to be plagued by infinite lists. When you ask a question that requires it to write a list, it most often keeps adding bullet points or enumeration. I am wondering whether this is a result of using chatty GPT-4 as DPO preferences. Any thoughts?
- 1 reply
s-conia
authored
a
paper
8 months ago
PSaiml
authored
a
paper
8 months ago
BramVanroy
posted
an
update
8 months ago
Post
2286
🥳 New license for datasets: Apache 2.0!
I have been struggling mentally for many months now with the OpenAI terms of use that indicate that their model outputs cannot be used to build "competing models". This leads to many questions:
- what is the definition of competing? Is it the same as "commercial"?
- since this is part of the terms of use between OpenAI and the API user, can a third party still use the generated dataset to build competing models?
- are such restrictions even legal in the first place?
Trying to "follow the rules" as much as possible despite wanting to be as open as possible, I kept releasing my datasets under non-commercial licenses (which are too restrictive anyhow - nothing should prevent you from using the data in non-LM commercial settings), just like models trained on these datasets. This has put me at a competitive disadvantage compared to creators who do not follow the same approach and release their data/models on apache 2.0 despite the OpenAI "restrictions". Moreover, I fear (https://twitter.com/BramVanroy/status/1780220420316164246) that my approach blocks adaptation of my data/models for (commercial) applications/integrations.
Thankfully @Rijgersberg noted that these OpenAI terms of use are NOT explicit in the Azure OpenAI API (https://twitter.com/E_Rijgersberg/status/1780308971762450725). Since my latest datasets were created via Azure, this comes as a relief. As far as I can tell after digging through Azure docs, this allows me to change all recent GPT4-generated datasets to apache 2.0! 🥳
- BramVanroy/ultrachat_200k_dutch
- BramVanroy/orca_dpo_pairs_dutch
- BramVanroy/ultra_feedback_dutch
- BramVanroy/ultra_feedback_dutch_cleaned
- BramVanroy/no_robots_dutch
I will have to mull over what I'll do for the older GPT3.5 datasets. What do you think that I should do?
I have been struggling mentally for many months now with the OpenAI terms of use that indicate that their model outputs cannot be used to build "competing models". This leads to many questions:
- what is the definition of competing? Is it the same as "commercial"?
- since this is part of the terms of use between OpenAI and the API user, can a third party still use the generated dataset to build competing models?
- are such restrictions even legal in the first place?
Trying to "follow the rules" as much as possible despite wanting to be as open as possible, I kept releasing my datasets under non-commercial licenses (which are too restrictive anyhow - nothing should prevent you from using the data in non-LM commercial settings), just like models trained on these datasets. This has put me at a competitive disadvantage compared to creators who do not follow the same approach and release their data/models on apache 2.0 despite the OpenAI "restrictions". Moreover, I fear (https://twitter.com/BramVanroy/status/1780220420316164246) that my approach blocks adaptation of my data/models for (commercial) applications/integrations.
Thankfully @Rijgersberg noted that these OpenAI terms of use are NOT explicit in the Azure OpenAI API (https://twitter.com/E_Rijgersberg/status/1780308971762450725). Since my latest datasets were created via Azure, this comes as a relief. As far as I can tell after digging through Azure docs, this allows me to change all recent GPT4-generated datasets to apache 2.0! 🥳
- BramVanroy/ultrachat_200k_dutch
- BramVanroy/orca_dpo_pairs_dutch
- BramVanroy/ultra_feedback_dutch
- BramVanroy/ultra_feedback_dutch_cleaned
- BramVanroy/no_robots_dutch
I will have to mull over what I'll do for the older GPT3.5 datasets. What do you think that I should do?
-
- 9 replies
BramVanroy
posted
an
update
9 months ago
Post
2440
🎈 LLM Benchmarks Update!
**tl;dr: do not depend on benchmark leaderboards to choose your "chatbot" model! (Especially for non-English languages.)**
First of all, I'm discontinuing the Open #Dutch #LLM Leaderboard (https://lnkd.in/eFnsaFR6). It will stay online for now, but I urge the use of the ScandEval leaderboard instead (https://scandeval.com/dutch-nlg/) by @saattrupdan . It contains more tasks, has better reproducibility and statistics (CI) and a flexible back-end library (
An important note here is that benchmark leaderboards are not a golden truth. Especially evaluating generative models is hard. You run into issues like prompt engineering (and sensitivity of models to one or other prompt), structured output generation, and - quite simply - "how to automatically evaluate open-ended generation".
💡 Another important but under-discussed facet is the discrepancy between models' capability of understanding vs. generating *in different languages* (so the NLU part of NLG benchmarking). In other words: some of the listed models score really well on, e.g., MCQ benchmarks but are not suitable to use as DUTCH chat bots. Interestingly, some of these models seem to understand questions in Dutch and are able to pick the right answer (because they have good knowledge or reasoning skills), but generating fluent and grammatical Dutch is something else entirely! This is perhaps also true for humans: it's easier to sort-of grasp the meaning of a new language and answer with "Yes" or "No", but answering fluently in the language is much harder! Yet, your language production fluency does not necessarily say anything about your knowledge and reasoning skills.
Hopefully we can get a chat arena for Dutch some day - user feedback is the most powerful metric!
**tl;dr: do not depend on benchmark leaderboards to choose your "chatbot" model! (Especially for non-English languages.)**
First of all, I'm discontinuing the Open #Dutch #LLM Leaderboard (https://lnkd.in/eFnsaFR6). It will stay online for now, but I urge the use of the ScandEval leaderboard instead (https://scandeval.com/dutch-nlg/) by @saattrupdan . It contains more tasks, has better reproducibility and statistics (CI) and a flexible back-end library (
scandeval) to run your own benchmarks with. As part of project "Leesplank" (with Michiel Buisman and Maarten Lens-FitzGerald) we recently added GPT-4-1106-preview scores to add a good "target" to the leaderboard.An important note here is that benchmark leaderboards are not a golden truth. Especially evaluating generative models is hard. You run into issues like prompt engineering (and sensitivity of models to one or other prompt), structured output generation, and - quite simply - "how to automatically evaluate open-ended generation".
💡 Another important but under-discussed facet is the discrepancy between models' capability of understanding vs. generating *in different languages* (so the NLU part of NLG benchmarking). In other words: some of the listed models score really well on, e.g., MCQ benchmarks but are not suitable to use as DUTCH chat bots. Interestingly, some of these models seem to understand questions in Dutch and are able to pick the right answer (because they have good knowledge or reasoning skills), but generating fluent and grammatical Dutch is something else entirely! This is perhaps also true for humans: it's easier to sort-of grasp the meaning of a new language and answer with "Yes" or "No", but answering fluently in the language is much harder! Yet, your language production fluency does not necessarily say anything about your knowledge and reasoning skills.
Hopefully we can get a chat arena for Dutch some day - user feedback is the most powerful metric!
-
- 3 replies
BramVanroy
posted
an
update
9 months ago
Post
2392
Does anyone have experience with finetuning Gemma? Even the 2B variant feels more memory heavy than mistral 7B. I know that its vocabulary is much larger (250k) but I'm a bit surprised that the max batch size that I can get in an A100 80GB is only 2 whereas I could fit 4 with mistral 7B - even though Gemma is much smaller except for the embedding layer. Both runs were using FA, same sequence length, same deepspeed zero 3 settings. Oh and yes I'm using the most recent hot fix of transformers that solves a memory issue with Gemma and others.
Any prior experience that you can share or suggestions to improve throughout?
Any prior experience that you can share or suggestions to improve throughout?
-
- 4 replies
BramVanroy
posted
an
update
9 months ago
Post
1725
🖴 The HPLT monolingual dataset has a new home!
After being in touch with HPLT folks, I've transfered the data to their org. That only makes sense. You can find it below.
HPLT/hplt_monolingual_v1_2
After being in touch with HPLT folks, I've transfered the data to their org. That only makes sense. You can find it below.
HPLT/hplt_monolingual_v1_2
-
- 3 replies
BramVanroy
posted
an
update
10 months ago
Post
🗄️ Massive data release on the HF Hub for 75 languages!
https://huggingface.co/datasets/BramVanroy/hplt_monolingual_v1_2
In December of last year, HPLT (https://hplt-project.org/) released version 1.2 of their dataset. It covers web-crawled data of 75 languages!, in the raw format as well as deduplicated and cleaned sections. In total, we're talking about over 40TB of data! This data was already accessible via their website but I figured the accessibility could be improved by an integration with Hugging Face tooling. 🤗 So I added the dataset here to the Hugging Face hub, enabing direct use in your conventional training pipelines for LLMs or other language technologies. The data will automatically be downloaded and optimised with just one line of code:
Let's use this big blob of data to build something awesome in our languages! 🥳
https://huggingface.co/datasets/BramVanroy/hplt_monolingual_v1_2
In December of last year, HPLT (https://hplt-project.org/) released version 1.2 of their dataset. It covers web-crawled data of 75 languages!, in the raw format as well as deduplicated and cleaned sections. In total, we're talking about over 40TB of data! This data was already accessible via their website but I figured the accessibility could be improved by an integration with Hugging Face tooling. 🤗 So I added the dataset here to the Hugging Face hub, enabing direct use in your conventional training pipelines for LLMs or other language technologies. The data will automatically be downloaded and optimised with just one line of code:
load_dataset("BramVanroy/hplt_mono_v1_2", "nl_cleaned")Let's use this big blob of data to build something awesome in our languages! 🥳
- 1 reply