NVIDIA Technical Blog | News and tutorials for developers, data ...
How do models represent style, and how can we more precisely extract and steer it?
A commonly requested feature in almost any LLM-based writing application is “I want the AI to respond in my style of writing,” or “I want the AI to adhere to this style guide.” Aside from costly and complicated multi-stage finetuning processes like Anthropic’s RL with... See more
A commonly requested feature in almost any LLM-based writing application is “I want the AI to respond in my style of writing,” or “I want the AI to adhere to this style guide.” Aside from costly and complicated multi-stage finetuning processes like Anthropic’s RL with... See more
Shortwave — rajhesh.panchanadhan@gmail.com [Gmail alternative]
Nicolay Gerold added
To address these concerns, several of the leading model companies are working on improved steering —a way to place better controls on LLM outputs—to focus model outputs and help models better understand and execute on complex user demands.
Sarah Wang • The Next Token of Progress: 4 Unlocks on the Generative AI Horizon
Darren LI added
in models with between 2 and 7 billion parameters, new capabilities emerge such as the ability to generate different creative text in formats like poems, code, scripts, musical pieces, emails, and letters, and to answer even open-ended and challenging questions in an informative way.
Ben Auffarth • Generative AI with LangChain: Build large language model (LLM) apps with Python, ChatGPT, and other LLMs
On the one hand, quality data still matters. A lot of focus on LLM improvement is on model and dataset size. There’s some early evidence that LLMs can be greatly influenced by the data quality they are trained with. WizardLM, TinyStories, and phi-1 are some examples. Likewise, RLHF datasets also matter.
On the other hand, ~100 data points is enough ... See more
On the other hand, ~100 data points is enough ... See more
Is Data Still a Moat?
Nicolay Gerold added
𝗺𝗲𝘁𝗵𝗼𝗱𝘀 𝗼𝗳 𝗳𝗶𝗻𝗲-𝘁𝘂𝗻𝗶𝗻𝗴 𝗮𝗻 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲 𝗟𝗟𝗠 𝗲𝘅𝗶𝘀t ↓
- 𝘊𝘰𝘯𝘵𝘪𝘯𝘶𝘦𝘥 𝘱𝘳𝘦-𝘵𝘳𝘢𝘪𝘯𝘪𝘯𝘨: utilize domain-specific data to apply the same pre-training process (next token prediction) on the pre-trained (base) model
- 𝘐𝘯𝘴𝘵𝘳𝘶𝘤𝘵𝘪𝘰𝘯 𝘧𝘪𝘯𝘦-𝘵𝘶𝘯𝘪𝘯𝘨: the pre-trained (base) model is fine-tuned on ... See more
- 𝘊𝘰𝘯𝘵𝘪𝘯𝘶𝘦𝘥 𝘱𝘳𝘦-𝘵𝘳𝘢𝘪𝘯𝘪𝘯𝘨: utilize domain-specific data to apply the same pre-training process (next token prediction) on the pre-trained (base) model
- 𝘐𝘯𝘴𝘵𝘳𝘶𝘤𝘵𝘪𝘰𝘯 𝘧𝘪𝘯𝘦-𝘵𝘶𝘯𝘪𝘯𝘨: the pre-trained (base) model is fine-tuned on ... See more
Shortwave — rajhesh.panchanadhan@gmail.com [Gmail alternative]
Nicolay Gerold added
Phi-1.5
Phi-1.5 is a "small" 1.3 billion parameter LLM with an impressive performance for its size.
Annotated figures from the Textbooks Is All You Need II paper
How does this small model accomplish such a good performance? The secret ingredient seems to be the high-quality data.
The pretraining is based on the Textbooks Is All You Need approach that... See more
Phi-1.5 is a "small" 1.3 billion parameter LLM with an impressive performance for its size.
Annotated figures from the Textbooks Is All You Need II paper
How does this small model accomplish such a good performance? The secret ingredient seems to be the high-quality data.
The pretraining is based on the Textbooks Is All You Need approach that... See more
Sebastian Raschka • Ahead of AI #12: LLM Businesses and Busyness
Nicolay Gerold added
The authors hypothesize that the model gains instruction following capabilities without being instruction finetuning, which is an interesting observation.
The model may have unintentionally been trained using benchmark datasets (mirrors test cases, but fails when format changes).
AgentTuning: Enabling Generalized Agent Abilities For LLMs
🤗 Model (AgentLM-70B) • 🤗 Dataset (AgentInstruct) • 📃 Paper • 🌐 Project Page
中文版(Chinese)
AgentTuning represents the very first attempt to instruction-tune LLMs using interaction trajectories across multiple agent tasks. Evaluation results indicate that AgentTuning enables the agent capa... See more
🤗 Model (AgentLM-70B) • 🤗 Dataset (AgentInstruct) • 📃 Paper • 🌐 Project Page
中文版(Chinese)
AgentTuning represents the very first attempt to instruction-tune LLMs using interaction trajectories across multiple agent tasks. Evaluation results indicate that AgentTuning enables the agent capa... See more
THUDM • GitHub - THUDM/AgentTuning: AgentTuning: Enabling Generalized Agent Abilities for LLMs
Nicolay Gerold added
Darren LI and added