You can run prompts to rough the new Llama 3 models on Replicate: https://
replicate.com/meta/meta-llam
a-3-70b-instruct
… for 70b-instruct and https://
replicate.com/meta/meta-llam
a-3-8b-instruct
… for 8b-instruct
LLMS
-
Llama 3 Models Now Available on Replicate Platform
By
–
-
Scaling Laws Evolution: 8B Model Trained on Fifteen Trillion Tokens
By
–
you're telling me an 8B param model was trained on fifteen trillion tokens? i didn't even know there was that much text in the world really interesting to see how scaling laws have changed best practices; GPT-3 was 175 billion params and trained on a paltry 300 billion tokens
-
Meta Llama 3 Now Available on Databricks Model Serving
By
–
Meta Llama 3 models are being rolled out across all Databricks Model Serving regions over the next few days. Once available, they can be accessed via the UI, API, or SQL interfaces.
-
LangSmith Evaluations: Summary Evaluators for Datasets
By
–
LangSmith Evaluations: Summary Evaluators for Datasets
— LangChain (@LangChain) 18 avril 2024
Evaluations can accelerate LLM app development, but it can be challenging to get started. We've kicked off a new video series focused on evaluations in LangSmith.
This is the 11th video in our series that shows how… pic.twitter.com/kaCIs9z9daLangSmith Evaluations: Summary Evaluators for Datasets Evaluations can accelerate LLM app development, but it can be challenging to get started. We've kicked off a new video series focused on evaluations in LangSmith. This is the 11th video in our series that shows how
-
LLaMA 3 Sequence Length Limitations Community Extensions
By
–
The main issue with the LLaMA 3 models is the sequence length… currently 8k. I'm confident the community will extend this pretty quickly.
-
Four GPT-4 Class Models Released: What’s Next for AI Development
By
–
We went from only one GPT-4 class model to four this year. Next move is OpenAI’s, which will tell us what the future looks like- plateauing technology or continued rapid development.
-
Open-Source GPT-4-Level Models Now Freely Accessible Globally
By
–
We're entering a new world where GPT-4-level models are open-source and freely accessible. Absolutely massive.
-
Training 400B+ Parameter Model to Outperform GPT-4
By
–
They're training a 400B+ parameter version that should outperform GPT-4.
-
ChatGPT vocabulary bias traced to Nigerian RLHF annotation workers
By
–
Love this theory by @alexhern that the reason ChatGPT uses words like "delve" a lot is that OpenAI outsource a lot of their RLHF annotation to workers in Nigeria, and Nigerian formal English has a slightly different vocabulary