Philipp Schmid
Nürnberg, Bayern, Deutschland
111.237 Follower:innen
500+ Kontakte
Nürnberg, Bayern, Deutschland
111.237 Follower:innen
500+ Kontakte
Aktivitäten
-
A few days ago a new method for generating preference data was released. 𝗖𝗼𝗻𝘁𝗿𝗮𝘀𝘁𝗶𝘃𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗔𝗜 𝗥𝗲𝘃𝗶𝘀𝗶𝗼𝗻…
A few days ago a new method for generating preference data was released. 𝗖𝗼𝗻𝘁𝗿𝗮𝘀𝘁𝗶𝘃𝗲 𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗔𝗜 𝗥𝗲𝘃𝗶𝘀𝗶𝗼𝗻…
Beliebt bei Philipp Schmid
-
The most comprehensive overview of LLM-as-a-Judge! READ IT! 🚨 "Evaluating the Effectiveness of LLM-Evaluators (aka LLM-as-Judge)” summarizes and…
The most comprehensive overview of LLM-as-a-Judge! READ IT! 🚨 "Evaluating the Effectiveness of LLM-Evaluators (aka LLM-as-Judge)” summarizes and…
Geteilt von Philipp Schmid
-
LLM pricing comparison space update with new prices for OpenAI, Google DeepMind, and new offerings from Cerebras Systems and AI21 Labs. 👀 > OpenAI…
LLM pricing comparison space update with new prices for OpenAI, Google DeepMind, and new offerings from Cerebras Systems and AI21 Labs. 👀 > OpenAI…
Geteilt von Philipp Schmid
Berufserfahrung
Ausbildung
Bescheinigungen und Zertifikate
Weitere Aktivitäten von Philipp Schmid
-
Dell Enterprise Hub now supports Kubernetes! Deploying the world’s largest and most capable openly available foundation model - Llama- 3.1 405B model…
Dell Enterprise Hub now supports Kubernetes! Deploying the world’s largest and most capable openly available foundation model - Llama- 3.1 405B model…
Beliebt bei Philipp Schmid
-
Announcing “Cloud AI Tuesdays”. 🚀 Every Tuesday, we will share detailed examples of how to build AI with open models in the Cloud (Google Cloud…
Announcing “Cloud AI Tuesdays”. 🚀 Every Tuesday, we will share detailed examples of how to build AI with open models in the Cloud (Google Cloud…
Geteilt von Philipp Schmid
-
Hugging Face Deep Learning Containers (DLCs) on Google Cloud Accelerating Machine Learning Hugging Face has recently contributed significantly to…
Hugging Face Deep Learning Containers (DLCs) on Google Cloud Accelerating Machine Learning Hugging Face has recently contributed significantly to…
Beliebt bei Philipp Schmid
-
Perform complex, multi-step reasoning tasks in dynamic, interactive environments by combining Monte Carlo Tree Search (MCTS) with self-critique and…
Perform complex, multi-step reasoning tasks in dynamic, interactive environments by combining Monte Carlo Tree Search (MCTS) with self-critique and…
Geteilt von Philipp Schmid
-
Last Week in AI was on fire! 🔥 From serverless LLMs to long-form content generation, here are the highlights: > Meta Llama 3 405B now available…
Last Week in AI was on fire! 🔥 From serverless LLMs to long-form content generation, here are the highlights: > Meta Llama 3 405B now available…
Geteilt von Philipp Schmid
-
We are excited to release the Hugging Face Deep Learning Containers (DLCs) for Google Cloud Repository! 🚀 The Google Cloud Containers provide public…
We are excited to release the Hugging Face Deep Learning Containers (DLCs) for Google Cloud Repository! 🚀 The Google Cloud Containers provide public…
Geteilt von Philipp Schmid
-
"AI is not making any progress"? Look closer. 🙄 OpenAI GPT-4 level models got 240x cheaper in just 2 years! AI progress isn't linear and is just…
"AI is not making any progress"? Look closer. 🙄 OpenAI GPT-4 level models got 240x cheaper in just 2 years! AI progress isn't linear and is just…
Geteilt von Philipp Schmid
-
New Long-output (> 2,000 words) open Dataset! Current LLMs can handle long inputs (100,000+ tokens) but struggle to generate outputs longer than…
New Long-output (> 2,000 words) open Dataset! Current LLMs can handle long inputs (100,000+ tokens) but struggle to generate outputs longer than…
Geteilt von Philipp Schmid
-
This week, Microsoft released 3 new Phi-3.5 models with Mini, Vision, and MoE. Did you know they can be converted to the Meta Llama architecture? 👀…
This week, Microsoft released 3 new Phi-3.5 models with Mini, Vision, and MoE. Did you know they can be converted to the Meta Llama architecture? 👀…
Geteilt von Philipp Schmid
-
Big Update! 20% higher throughput and 60% memory reduction for multi-GPU fine-tuning with Hugging Face Transformers! 🤯 The LLM research team from…
Big Update! 20% higher throughput and 60% memory reduction for multi-GPU fine-tuning with Hugging Face Transformers! 🤯 The LLM research team from…
Geteilt von Philipp Schmid
-
New promising RLHF and synthetic data generation method! 👀 APO and CLAIR improve the performance of Llama 3.1 8B by 7.45% on MixEval Hard…
New promising RLHF and synthetic data generation method! 👀 APO and CLAIR improve the performance of Llama 3.1 8B by 7.45% on MixEval Hard…
Geteilt von Philipp Schmid
-
Turns out small talk is more difficult than expected for LLM Chat models. Standard SFT datasets often fail at simple greetings and go completely…
Turns out small talk is more difficult than expected for LLM Chat models. Standard SFT datasets often fail at simple greetings and go completely…
Geteilt von Philipp Schmid
-
Mixture of Experts of different sizes? 🤔 HMoE: Heterogeneous Mixture of Experts for Language Modeling introduces the concept of using experts of…
Mixture of Experts of different sizes? 🤔 HMoE: Heterogeneous Mixture of Experts for Language Modeling introduces the concept of using experts of…
Geteilt von Philipp Schmid
Weitere ähnliche Profile
Weitere Mitglieder namens Philipp Schmid in Deutschland
-
Philipp Schmid
Vice President Human Resources bei E.ON Energie Deutschland GmbH
-
Philipp Schmid
-
Philipp Schmid
-
Philipp Schmid
EY | DKB | B.A. Betriebswirtschaftslehre
Es gibt auf LinkedIn 84 weitere Personen namens Philipp Schmid, die sich in Deutschland befinden.
Weitere Mitglieder anzeigen, die Philipp Schmid heißen