@TimilsinaBml05 Profile picture

Bimal Timilsina

@TimilsinaBml05

ML Engineer | I write about Python and Machine Learning |

Joined July 2016
Similar User
सोझो मान्छे photo

@pa_amrit

Rd mahesh photo

@raid_mahesh

Smn photo

@script_smn07

🇳🇵ᴅɪɴᴇꜱʜ🇳🇵 photo

@its_me_dns

सु र ज photo

@SurajKandel12

The best use case of removing distraction feature of Safari is to remove Youtube shorts section in homepage. Works like a charm. Try it.


Can we say the reason OpenAI leads the lmsys leaderboard is since they train on the user prompts? They say they don’t train data from APIs but what if they do? If lmsys uses apis then they have access to all the data on which they can train new models and the results we all know.


Bimal Timilsina Reposted

Dear Happy Dressing Room, वर्ल्ड कप सकिएको छ, तर हाम्रो यात्रा अझै बाँकी छ। ती तीन नजिकका खेलहरूमा, हामी विजयको नजिकै पुगेका थियौं। नतिजा हाम्रो पक्षमा नभए पनि, हामीले हाम्रो सम्भावना र लड्ने आत्मा देखायौं। हामी ब्याटिङ युनिटको रूपमा जिम्मेवारी र जवाफदेहीता लिनुपर्छ। हामी…

Tweet Image 1

Bimal Timilsina Reposted

Jay 🇳🇵

Tweet Image 1

Bimal Timilsina Reposted

One of the great wins in Nepali cricketing history over a very strong West Indies side in front of 20,000+ with a truly unforgettable captain's innings from Rohit Paudel with 112 off 54 balls as Nepal chase down 204 to go 1-0 up in the series!!🇳🇵 Over 265,000 people online too!

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Really Poor Camerawork by @KantipurTVHD . So sad to see the this level of negligence in the quality of live in this historic series. Atleast Use Good Cameraman please! @CricketNep @paras77 #NepvsWI #NEPvsWIA


Bimal Timilsina Reposted

Finally! What we all have been waiting for - Llama 3 weights are on @huggingface: huggingface.co/collections/me…


Bimal Timilsina Reposted

Introducing `claude-researcher` 📈 A powerful Claude 3 research agent that delivers thorough reports in record time. Just provide an topic, and a chain of AIs with **access to Google** will generate an incredibly comprehensive report for you. And it's open-source!


Bimal Timilsina Reposted

Training LLMs can be much cheaper than previously thought. 0.1 million USD is sufficient for training LLaMA2-level LLMs🤯 While @OpenAI and @Meta use billions of dollars to train theirs, you can also train yours with much less money. Introducing our open-source project JetMoE:…

Tweet Image 1

Bimal Timilsina Reposted

Outperforming Llama 2 7B and LLaMA 13B with <$100K. 8B sparse MoE model, with 2B activated parameters. huggingface.co/jetmoe/jetmoe-…


Bimal Timilsina Reposted

Instructor is awesome! You can use Outlines to do the same thing with open source models github.com/outlines-dev/o…


Bimal Timilsina Reposted

I used to beg LLMs for structured outputs. Most of the time, they understood the job and returned valid JSONs. However, around ~5% of the time, they didn't, and I had to write glue code to avoid breaking downstream modules of my system. Then, I found @jxnlco's instructor.…

Tweet Image 1
Tweet Image 2

Bimal Timilsina Reposted

What a week! My author copies have just arrived. This is exciting! Thanks everyone for all the encouragement to write this book! If you've been on the hunt for the next resource to improve your understanding of machine learning and AI concepts, "Machine Learning Q & AI" with…

Tweet Image 1

Bimal Timilsina Reposted

I’m gonna give 10 random people that repost this and follow me $25,000 for fun (the $250,000 my X video made) I’ll pick the winners in 72 hours


Bimal Timilsina Reposted

For an unknown limited time (probably until some crashing bug), I am hosting Stable Beluga 2 70B (a tune of Llama 2 70B) which is a little more "guide-able" than Llama 2 via prompting. It's a streaming-chat Gradio app hosted on @LambdaAPI cloud. http://209.20.159.223:7860/

Tweet Image 1

Bimal Timilsina Reposted

GPT-4 is getting worse over time, not better. Many people have reported noticing a significant degradation in the quality of the model responses, but so far, it was all anecdotal. But now we know. At least one study shows how the June version of GPT-4 is objectively worse than…

Tweet Image 1

Bimal Timilsina Reposted

JUST IN: Meta AI introduces Llama 2 It’s been ~4 months since Llama was launched and took the community by storm. This new release brings a few new things. Here is the breakdown: - The release involves a partnership with Microsoft - Available for research and commercial use -…

Tweet Image 1

Bimal Timilsina Reposted

✨NEW LAUNCH! LLaMA2 chat API & open-source playground💫: We're releasing tools that make it easy to test @meta's latest LLM & add it to your own app with @replicatehq Playground: llama2.ai Live chat API here: replicate.com/a16z-infra/lla… Repos & instructions below:


Bimal Timilsina Reposted

Huge day indeed for AI and LLMs, congrats to Meta 👏 This is now the most capable LLM available directly as weights to anyone from researchers to companies. The models look quite strong, e.g. Table 4 in the paper: MMLU is good to look at, the 70B model is just below GPT-3.5. But…

Tweet Image 1

This is huge: Llama-v2 is open source, with a license that authorizes commercial use! This is going to change the landscape of the LLM market. Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers Pretrained and fine-tuned…



Bimal Timilsina Reposted

This is huge: Llama-v2 is open source, with a license that authorizes commercial use! This is going to change the landscape of the LLM market. Llama-v2 is available on Microsoft Azure and will be available on AWS, Hugging Face and other providers Pretrained and fine-tuned…


Loading...

Something went wrong.


Something went wrong.