grontown
ad-llama
grontown | ad-llama | |
---|---|---|
3 | 6 | |
119 | 47 | |
- | - | |
8.7 | 8.9 | |
8 months ago | about 1 month ago | |
TypeScript | TypeScript | |
- | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
grontown
- FLaNK Stack for 25 September 2023
- eastworld - an open-source framework for generative game agents
-
Show HN: A murder mystery game built on an open-source gen-AI agent framework
Hey HN,
Michael and Scott here. We’re open-sourcing an interactive murder mystery featuring LLM-driven character agents. Solve the mystery by finding clues, taking notes, and interrogating agents. They all have distinct motives, personality, and can impact the game in different ways (attacking you, running away, etc). Try it out, it’s pretty fun!
We’re also open-sourcing the framework that we used to make and refine the agents. The goal is to create an intuitive interface for storytellers to create, debug, and test game agents. We then take those game agents and expose an API beyond just chat - such as actions, player guardrails, emotional queries, etc.
We’re not done yet - there are a lot more features coming on the way: scenario-based agent evals, agent-storyline consistency management, automatic agent generation, etc.
We would love to hear your feedback.
Thanks!
[0] https://github.com/mluogh/grontown
ad-llama
- Show HN: A murder mystery game built on an open-source gen-AI agent framework
-
Guidance: A guidance language for controlling large language models
I took a stab at making something[1] like guidance - I'm not sure exactly how guidance does it (and I'm also really curious how it would work with chat api's) but here's how my solution works.
Each expression becomes a new inference request, so it's not a single inference pass. Because each subsequent pass includes the previously inferenced text, the LLM ends up doing a lot of prefill and less decode. You only decode as much as you actually inference, the repeated passes only end up costing more in prefill (which tend to be much faster tok/s).
To work with chat tuned instruction models, you can basically still treat it as a completion model. I provide the previously completed inference text as a partially completed assistant response, e.g. with llama 2 it goes after [/INST]. You can add a bit of instruction for each inference expression which gets added to the [INST]. This approach lets you start off the inference with `{ "someField": "` for example to guarantee (at least the start of) a json response and allow you to add a little bit of instruction or context just for that field.
I didn't even try with openai api's since afaict you can't provide a partial assistant response for it to continue from. Even if you were to request a single token at a time and use logit_bias for biased sampling, I don't see how you can get it to continue a partially completed inference.
[1] https://github.com/gsuuon/ad-llama
-
Simulating History with ChatGPT
Can you point me to some text-adventure engines? I'm hacking on an in-browser local llm structured inference library[1] and am trying to put together a text game demo[2] for it. It didn't even occur to me that text-adventure game engines exist, I was apparently re-inventing the wheel.
[1] https://github.com/gsuuon/ad-llama
[2] https://ad-llama.vercel.app/murder/
-
Ask HN: Which programming language to learn in AI era?
Yup, I'm building a library that runs LLM's in browser with tagged template literals: https://github.com/gsuuon/ad-llama
I think it has fundamental DX benefits over python for complex prompt chaining (or I wouldn't be building it!) Even still -- if their focus is purely on AI, python is still the better choice starting from scratch. The python AI ecosystem has many more libraries, stack overflow answers, tutorials, etc available.
-
Show HN: LLMs can generate valid JSON 100% of the time
Generating an FSM over the vocabulary is a really interesting approach to guided sampling! I'm hacking on a structured inference library (https://github.com/gsuuon/ad-llama) - I also tried to add a vocab preprocessing step to generate a valid tokens mask (just with regex or static strings initially) but discovered that doing so would cause unlikely / unnatural tokens to be masked rather than the token which represents the natural encoding given the existing sampled tokens.
Given the stateful nature of tokenizers, I decided that trying to preprocess the individual token ids was a losing battle. Even in the simple case of whitespace - tokenizer merges can really screw up generating a static mask, e.g. we expect a space next, but a token decodes to 'foo', but is actually a '_foo' and would've decoded with a whitespace if it were following a valid pair. When I go to construct the static vocab mask, it would then end up matching against 'foo' instead of ' foo'.
How did you work around this for the FSM approach? Does it somehow include information about merges / whitespace / tokenizer statefulness?
What are some alternatives?
eastworld - Framework for Generative Agents in Games
llm - Access large language models from the command-line
FLaNK-SaoPauloBrazil - FLaNK-SaoPauloBrazil
llm-mlc - LLM plugin for running models using MLC
SeaGOAT - local-first semantic code search engine
every-breath-you-take - Heart Rate Variability Training with the Polar H10 Monitor
hof - Framework that joins data models, schemas, code generation, and a task engine. Language and technology agnostic.
GlobalMLBuildingFootprints - Worldwide building footprints derived from satellite imagery
outlines - Structured Text Generation
CML_AMP_Finetune_Foundation_Model_Multiple_Tasks
api - Structured LLM APIs