When I want to type a special non-English character like Ø, on my Android phone I simply long press on "o" on the keyboard and it brings up every variation of the letter o you could think of.

On Windows, the best you get is "please memorise this giant list of four number codes". If I'm going to have to google the code I'm probably just going to copy the letter and paste it into whatever I'm writing.

Why not add them to the damn emoji menu (Windows key + .)

https://theconversation.com/us-military-plans-to-unleash-thousands-of-autonomous-war-robots-over-next-two-years-212444

Got a notification from LinkedIn saying "You're one of the few experts who have been invited to collaborate on ..." I got curious and opened up the link.

Apparently, now instead of professional writers being paid to pen down their, usually, cohesive & authentic views, LinkedIn is trying out the idea of generating content using an LLM and then asking for free editorial services from users in exchange for "badges" 🤯 🤦‍♂️

This is cheap IMO. Even for LinkedIn.

What's happened to the "content team" at LinkedIn!?

A much needed free course on Foundations of Humane Technology for the era of rapid technological changes and advances.

https://www.humanetech.com/course

cross-posted from: https://lemmy.nz/post/1310373

Ventoy is an opensource tool to create a universal bootable USB drive for ISOs and other image files. With Ventoy, you don't need to format the disk over and over to create a bootable USB for different images, you just need to copy the image files to the USB drive and boot them directly via a dynamic menu.

New in v1.0.95:

• Drag to resize Ventoy2Disk.exe dialog width.
• Fix a bug when booting veket_24.
• Fix a bug when booting the latest UOS server ISO.
• New distro support: vanilladpup
• New distro support: FydeOS 17
• languages.json update

cross-posted from: https://kbin.social/m/workreform@lemmy.world/t/401046

Emotion recognition systems are finding growing use, from monitoring customer responses to ads to scanning for ‘distressed’ women in danger.

In July, Lockheed Martin completed the build of NASA’s X-59 test aircraft, which is designed to turn sonic booms into mere thumps, in the hope of making overland supersonic flight a possibility. Ground tests and a first test flight are planned for later in the year. NASA aims to have enough data to hand over to US regulators in 2027.

cross-posted from: https://lemmy.world/post/3879861

Beating GPT-4 on HumanEval with a Fine-Tuned CodeLlama-34B

Hello everyone! This post marks an exciting moment for !fosai@lemmy.world and everyone in the open-source large language model and AI community.

We appear to have a new contender on the block, a model apparently capable of surpassing OpenAI's state of the art ChatGPT-4 in coding evals (evaluations).

This is huge. Not too long ago I made an offhand comment on us catching up to GPT-4 within a year. I did not expect that prediction to end up being reality in half the time. Let's hope this isn't a one-off scenario and that we see a new wave of open-source models that begin to challenge OpenAI.

Buckle up, it's going to get interesting!

Here's some notes from the blog, which you should visit and read in its entirety:

• https://www.phind.com/blog/code-llama-beats-gpt4

Blog Post

We have fine-tuned CodeLlama-34B and CodeLlama-34B-Python on an internal Phind dataset that achieved 67.6% and 69.5% pass@1 on HumanEval, respectively. GPT-4 achieved 67% according to their official technical report in March. To ensure result validity, we applied OpenAI's decontamination methodology to our dataset.

The CodeLlama models released yesterday demonstrate impressive performance on HumanEval.

• CodeLlama-34B achieved 48.8% pass@1 on HumanEval
• CodeLlama-34B-Python achieved 53.7% pass@1 on HumanEval

We have fine-tuned both models on a proprietary dataset of ~80k high-quality programming problems and solutions. Instead of code completion examples, this dataset features instruction-answer pairs, setting it apart structurally from HumanEval. We trained the Phind models over two epochs, for a total of ~160k examples. LoRA was not used — both models underwent a native fine-tuning. We employed DeepSpeed ZeRO 3 and Flash Attention 2 to train these models in three hours using 32 A100-80GB GPUs, with a sequence length of 4096 tokens.

Furthermore, we applied OpenAI's decontamination methodology to our dataset to ensure valid results, and found no contaminated examples. 

The methodology is:

• For each evaluation example, we randomly sampled three substrings of 50 characters or used the entire example if it was fewer than 50 characters.
• A match was identified if any sampled substring was a substring of the processed training example.

For further insights on the decontamination methodology, please refer to Appendix C of OpenAI's technical report. Presented below are the pass@1 scores we achieved with our fine-tuned models:

• Phind-CodeLlama-34B-v1 achieved 67.6% pass@1 on HumanEval
• Phind-CodeLlama-34B-Python-v1 achieved 69.5% pass@1 on HumanEval

Download

We are releasing both models on Huggingface for verifiability and to bolster the open-source community. We welcome independent verification of results.

• https://huggingface.co/Phind/Phind-CodeLlama-34B-v1
• https://huggingface.co/Phind/Phind-CodeLlama-34B-Python-v1

If you get a chance to try either of these models out, let us know how it goes in the comments below!

If you found anything about this post interesting, consider subscribing to !fosai@lemmy.world.

Cheers to the power of open-source! May we continue the fight for optimization, efficiency, and performance.

@technology
The EU has just clamped down on big tech. Britain, take note https://mastodon.scot/@DrHannahGraham/110956248305740892