in the community: huggingface transformers, axolotl, unsloth, lit-gpt are popular options.
AI
-

Llama 3 Models Now Available in LangSmith Playground
By
–
Try out the latest Llama 3 8B and 70B models in LangSmith Playground today!
— LangChain (@LangChain) 18 avril 2024
Thanks to our friends at @FireworksAI_HQ! https://t.co/0d6H7X71r1 pic.twitter.com/lsC9Cjr3NYTry out the latest Llama 3 8B and 70B models in LangSmith Playground today! Thanks to our friends at @FireworksAI_HQ
! -
Serverless Fine-tuning and Serving for Llama3 on Predibase
By
–
Check out #serverless fine-tuning and serving for #Llama3 on @Predibase with our free trial! 🦙
— Predibase by Rubrik (@predibase) 18 avril 2024
🔥 Blazing fast serverless inference – #8b & #70b variants
🖌️ #Finetune in the UI or SDK via config
🌳 Start prompting your fine-tune instantly with #LoRAXhttps://t.co/L3Tv8r07P1 pic.twitter.com/GAOW88SfZzCheck out #serverless fine-tuning and serving for #Llama3 on @Predibase with our free trial! Blazing fast serverless inference – #8b & #70b variants #Finetune in the UI or SDK via config Start prompting your fine-tune instantly with #LoRAX https://
pbase.ai/3VZE37X -

Serverless Fine-tuning and Serving for Llama3 on Predibase
By
–
Check out #serverless fine-tuning and serving for #Llama3 on @Predibase with our free trial! 🦙
— Predibase by Rubrik (@predibase) 18 avril 2024
🔥 Blazing fast serverless inference – #8b & #70b variants
🖌️ #Finetune in the UI or SDK via config
🌳 Start prompting your fine-tune instantly with #LoRAXhttps://t.co/Rq8fsQwaZl pic.twitter.com/CYON9Wx25ZCheck out #serverless fine-tuning and serving for #Llama3 on @Predibase with our free trial! Blazing fast serverless inference – #8b & #70b variants #Finetune in the UI or SDK via config Start prompting your fine-tune instantly with #LoRAX https://
predibase.com/free-trial -
M3 Max 8B model optimization for multi-agent orchestration systems
By
–
I have an M3 Max with 64GB. You can use the 8B model if you don't have a lot of VRAM. What I realize is that the 8B model works really well as a subagent. As long as the Orchestrator and the Refiner are the 70GB models, the whole thing works really well.
-
Ollama Integration with Llama 3 Orchestration Works Well
By
–
Some considerations:
It works surprisingly well, not many changes needed besides, of course, the different way @ollama inference works. I just had to add some more info in the prompt for the orchestrator because Llama 3 loves to yap, lol. At one point, it forgot the goal and -

Meta Llama 3 Impresses Despite Smaller Model Size
By
–
Meta Llama 3 is very good, especially for such a small model. We can put in a multi-page prompt like our negotiation simulator (
https://
moreusefulthings.com/student-exerci
ses
…) & it is able to follow the complexity reasonably well. It doesn’t have the “smarts” of GPT-4 class, but impressive nonetheless. -
Maestro-Ollama: Local Llama 3 70B Agent Framework
By
–
Introducing Maestro-Ollama! 🦙
— Pietro Schirano (@skirano) 18 avril 2024
You can now harness the power of the Maestro framework entirely locally using Llama 3 70B via @ollama.
Let that sink in for a second, this is a model that outperforms Claude 3 Sonnet, operating as an agent, completely locally.
What a time! 🔥 pic.twitter.com/aC34Bd6F65Introducing Maestro-Ollama! You can now harness the power of the Maestro framework entirely locally using Llama 3 70B via @ollama
. Let that sink in for a second, this is a model that outperforms Claude 3 Sonnet, operating as an agent, completely locally. What a time! -
Fine-tuning Models Before Release for Better Self-Description
By
–
I agree with Jeremy here: we know that people are going to ask models about themselves here whether or not that actually makes sense to do so, so we should be able to fine-tune the model after initial training but before release to help it provide more useful answers there
-
RLHF Annotation Bias and Model Vocabulary Development
By
–
I don't know that RLHF would bias that kind of thing – my mental model is that annotators are shown two answers to the same prompt and asked which is "best", so if none of the test prompts happened to touch on the concept of a roadside kiosk that vocabulary wouldn't be affected