Llama-v2 is open source, with a license that authorizes commercial use! Llama 2 is a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters.
Fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Llama 2-Chat models outperform open-source chat models on most benchmarks, and based on human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models.
In this video, I will talk about the following
00:00:00 What is Llama 2?
00:04:40 How are Llama 2-Chat models trained?
00:06:25 How are Llama 2 models pretrained?
00:08:16 Safety Characteristics in Pretraining data
00:10:47 How do Llama 2 models perform?
00:13:33 How are Llama 2 models finetuned?
00:20:40 Llama 2 models finetuning: Reward Modeling for RLHF
00:27:52 System Message for Multi-Turn Consistency
00:30:52 How do Llama 2 Chat models perform?
00:34:54 How safe are Llama 2 Pretrained and Finetuned Models?
00:37:20 Context Distillation for Safety in RLHF finetuning
00:41:46 Tool Use Emergence in Llama 2-Chat
For more details, please look at https://ai.meta.com/research/publications/llama-2-open-foundation-and-fine-tuned-chat-models/
Llama 2: Open Foundation and Fine-Tuned Chat Models. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi and others. July 2023.
Fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Llama 2-Chat models outperform open-source chat models on most benchmarks, and based on human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models.
In this video, I will talk about the following
00:00:00 What is Llama 2?
00:04:40 How are Llama 2-Chat models trained?
00:06:25 How are Llama 2 models pretrained?
00:08:16 Safety Characteristics in Pretraining data
00:10:47 How do Llama 2 models perform?
00:13:33 How are Llama 2 models finetuned?
00:20:40 Llama 2 models finetuning: Reward Modeling for RLHF
00:27:52 System Message for Multi-Turn Consistency
00:30:52 How do Llama 2 Chat models perform?
00:34:54 How safe are Llama 2 Pretrained and Finetuned Models?
00:37:20 Context Distillation for Safety in RLHF finetuning
00:41:46 Tool Use Emergence in Llama 2-Chat
For more details, please look at https://ai.meta.com/research/publications/llama-2-open-foundation-and-fine-tuned-chat-models/
Llama 2: Open Foundation and Fine-Tuned Chat Models. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi and others. July 2023.