正在加载视频...

视频加载失败

"100 million words context window is already possible, which is roughly what a human hears in a lifetime. Inference support is the only bottleneck to achieve it. And AI Models actually do learn during the context window, without changing the weights." ~ Anthropic CEO Dario Amodei (On the 2nd...

631,729 次观看 • 10 个月前 •via X (Twitter)

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

Andrew Ng

200,950 次观看 • 1 年前

context engineering vs graph engineering. every few months the list gets a new word and everyone treats it as a replacement for the last one. these two are not on the same list. one decides what the model sees this turn, the other decides what exists at all. the cleanest way to tell them apart is to ask what a single unit of work looks like. > context engineering is the window the window opens empty, every single time. you assemble what goes in it. the prompt, the docs, the history, the tool results. the assembling is the work. the window only grows. it never shrinks on its own, so eventually something gets dropped. usually from the middle. usually without telling you. then the turn ends and the window is thrown away. not archived, thrown away. the next turn opens empty again and you re-explain what you already explained. good context engineering is knowing what to leave out, not what to pack in. the unit of work is one window. > graph engineering is the structure the same material arrives from the same sources. instead of packing it into a window, you pull entities out of it, resolve the duplicates into one node, and write typed edges between them. nothing here is stored as text you hope to find again. it is stored as a thing with a name and its connections to other things. when the turn ends, the graph is still there. the next turn does not start from zero. it starts by querying what already exists, and the query walks edges instead of guessing at similarity. good graph engineering is deciding what counts as the same thing twice. the unit of work is one relationship. > they are not alternatives the graph is what refills the window. context engineering decides what fits. graph engineering decides what there is to choose from. remove the graph and every session starts blind. remove the context work and the best structure in the world arrives as an unreadable dump. that also tells you which one broke. the answer drifted from what you actually said, or forgot something from this same session. that is the window. the answer is coherent but invents a connection that does not exist, or cannot join two facts it has clearly seen. that is the structure. people debug the prompt because the prompt is the easiest thing to edit. it keeps taking the blame for failures that live a layer down. save this - then read the full breakdown below

Hanako

19,160 次观看 • 24 天前

Dario Amodei just dismantled the biggest myth in the AI industry. Open source AI isn’t free. It never was. Amodei: “It’s not free. You have to run it on inference and someone has to make it fast on inference.” For decades, open source meant something real. It meant a teenager in a basement could download the same tools as a Fortune 500 company. Could read the code. Could modify it. Could build something that competed with the giants. That was genuine democratization. That actually happened. AI is different. Fundamentally. Physically. In ways the ideology hasn’t caught up to yet. Downloading the weights is the easy part. The part that actually costs something is turning the weights into a running system. Into responses. Into intelligence operating in real time at scale. That requires compute. Power. Infrastructure. The kind measured in billions of dollars and years of construction. Amodei: “These are big models. They’re hard to do inference on. Ultimately you have to host it on the cloud. The people who host it on the cloud do inference.” The open source debate was never about who owns the model. It was always about who owns the cloud. And Amodei goes further. When a competitor drops a new open model, he doesn’t ask whether it’s open or closed. He doesn’t care about the licensing. He doesn’t engage the ideology. Amodei: “I don’t think it mattered that DeepSeek is open source. I think I ask, is it a good model? Is it better than us at the things that matter? That’s the only thing that I care about.” That’s the ruthless clarity of someone actually trying to win. While the media debates licensing frameworks, Amodei is asking one question. Is it better. Everything else is a distraction. Amodei: “I don’t think open source works the same way in AI that it has worked in other areas. Here we can’t see inside the model.” This isn’t Linux. You can’t read it. You can’t fork it. You can’t understand it the way generations of developers understood the tools they inherited. You can download it. And then you need a data center to run it. The teenager in the basement who was supposed to be empowered by this revolution needs a billion dollars of infrastructure before the empowerment starts. The era of the basement coder rewriting civilization on a laptop is over. The future belongs to whoever commands the compute, owns the power grid, and can actually turn the intelligence on. Open weights without infrastructure isn’t democratization. It’s a promise the physics of the universe won’t let us keep.

Dustin

687,563 次观看 • 5 个月前

Gemini-1.5 Pro has its spotlight stolen today, and people are poking fun at Sora vs Google memes. Well, I think it's the biggest boost in LLM capability so far in 2024. v1.5's 10M token context (1) excels at retrieval; (2) generalizes zero-shot to extremely long instructions like full tutorials and codebases; and (3) works across modalities such as text, audio, and video. Here's a stunning example: v1.5 learns to translate from English to Kalamang purely in context, following a full linguistic manual at inference time. Kalamang is a language spoken by fewer than 200 speakers in western New Guinea. Gemini has never seen this language during training and is only provided with 500 pages of linguistic documentation, a dictionary, and ~400 parallel sentences in context. It basically acquires a sophisticated new skill in the neural activations, instead of gradient finetuning. I talked about the Myth of Context Length many times before: don't get too excited by claims of 1M or even 1B context tokens. LSTMs already achieved literally infinite context length 25 yrs ago! What truly matters is how well the model actually uses the context to solve real-world problems, and Gemini-1.5 has surpassed the SOTA with flying colors. The paper is also well-written with lots of solid quantitative analysis on in-context memorization and generalization. Paper: “Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context” Congrats to Jeff Dean Oriol Vinyals Sundar Pichai and team!

Jim Fan

278,517 次观看 • 2 年前