Sublime
An inspiration engine for ideas
The lack of acknowledgment of the threat of prompt injection attacks in this new Purple Llama initiative from Meta AI Is baffling to me
Guideline following Large Language Model for Information Extraction
Model Card for GoLLIE 34B
We present GoLLIE, a Large Language Model trained to follow annotation guidelines. GoLLIE outperforms previous approaches on zero-shot Information Extraction and allows the user to perform inferences with annotation schemas defined on the fly. Different from... See more
Model Card for GoLLIE 34B
We present GoLLIE, a Large Language Model trained to follow annotation guidelines. GoLLIE outperforms previous approaches on zero-shot Information Extraction and allows the user to perform inferences with annotation schemas defined on the fly. Different from... See more
HiTZ/GoLLIE-34B · Hugging Face
SGLang is a structured generation language designed for large language models (LLMs). It makes your interaction with LLMs faster and more controllable by co-designing the frontend language and the runtime system.
The core features of SGLang include:
The core features of SGLang include:
- A Flexible Front-End Language : This allows for easy programming of LLM applications with multiple ch
sgl-project • GitHub - sgl-project/sglang: SGLang is a structured generation language designed for large language models (LLMs). It makes your interaction with models faster and more controllable.
engineers continuously monitored features for and predictions from production models (Lg1, Md1, Lg3, Sm3, Md4, Sm6, Md6, Lg5, Lg6): Md1 discussed hard constraints for feature columns (e.g., bounds on values), Lg3 talked about monitoring completeness (i.e., fraction of non-null values) for features, Sm6 mentioned embedding their pipelines with "comm... See more
Shreya Shankar • "We Have No Idea How Models will Behave in Production until Production": How Engineers Operationalize Machine Learning.
Very recently, certain animal brains have begun to exhibit both generality of optimization power (producing an amazingly wide range of artifacts, in time scales too short for natural selection to play any significant role) and cumulative optimization power (artifacts of increasing complexity, as a result of skills passed on through language and wri
... See moreEliezer Yudkowsky • Rationality
pair-preference-model-LLaMA3-8B by RLHFlow: Really strong reward model, trained to take in two inputs at once, which is the top open reward model on RewardBench (beating one of Cohere’s).
DeepSeek-V2 by deepseek-ai (21B active, 236B total param.): Another strong MoE base model from the DeepSeek team. Some people are questioning the very high MMLU sc... See more
DeepSeek-V2 by deepseek-ai (21B active, 236B total param.): Another strong MoE base model from the DeepSeek team. Some people are questioning the very high MMLU sc... See more
Shortwave — rajhesh.panchanadhan@gmail.com [Gmail alternative]
projects are driven by democratized groups.1 Many find this notion of an alpha leader romantically appealing, believing that great design requires a tyrannical “Steve Jobs” at the helm to be successful.
Kritina Holden • Universal Principles of Design, Revised and Updated: 125 Ways to Enhance Usability, Influence Perception, Increase Appeal, Make Better Design Decisions, and Teach through Design
MIT Media Lab: David Rudnick
youtube.comFor example, if you ask a model to “return all active users in the last 7 days” it might hallucinate a `is_active` column, join to an `activity` table that doesn’t exist, or potentially get the wrong date (especially in leap years!).
We previously talked to Shreya Rajpal at Guardrails AI, which also supports Text2SQL enforcement. Their approach was ... See more
We previously talked to Shreya Rajpal at Guardrails AI, which also supports Text2SQL enforcement. Their approach was ... See more