rckrd

12 ms·

91.
AIM: AI Instant Messenger(chat.matt-rickard.com)
4 points by rckrd 3y ago|1 comments
92.
Context-Free Grammar Parsing with LLMs(matt-rickard.com)
14 points by rckrd 3y ago|0 comments
93.
Why Multi-Cloud Failed(matt-rickard.com)
3 points by rckrd 3y ago|0 comments
94.
Self-Hosted Compilers and Bootstrapped AI(matt-rickard.com)
2 points by rckrd 3y ago|0 comments
95.
Why Declarative Build Systems Aren’t Popular(matt-rickard.com)
1 points by rckrd 3y ago|1 comments
96.
OpenLM: OpenAI-compatible Python client to call any hosted LLM inference API(github.com)
1 points by rckrd 3y ago|0 comments
97.
r2d4/llm.ts: Call any LLM with a single API. Zero dependencies(github.com)
3 points by rckrd 3y ago|0 comments
98.
by rckrd 3y ago
There's probably a better API that wraps generate, but there's a bit more work than the logit mask. You have to go one token at a time, otherwise the masking becomes combinatoric rather than linear (two tokens at a time -- need to
99.
by rckrd 3y ago
(author here) That's interesting! Maybe there's a way to quantify the cumulative probability of the squashed tokens (i.e., if you constrain to 'true' and 'false', what's the distribution of the other token
100.
by rckrd 3y ago
Similar strategies with the logitsprocesor. It is a more generalized version that's not just constrained to JSON parsing, but any regex. JSONformer/clownfish try to parse the types syntactically. A regex is a better fit for a diff
101.
ReLLM: Exact Structure for Large Language Model Completions(matt-rickard.com)
26 points by rckrd 3y ago|13 comments
102.
The Ptrace Syscall(matt-rickard.com)
1 points by rckrd 3y ago|0 comments
103.
Sandbox Your Prompts(matt-rickard.com)
3 points by rckrd 3y ago|0 comments
104.
Jevons Paradox and LLMs(matt-rickard.com)
2 points by rckrd 3y ago|1 comments
105.
Synthetic Data from Compilers(matt-rickard.com)
1 points by rckrd 3y ago|0 comments
106.
Foundational Models Are Not Enough(matt-rickard.com)
3 points by rckrd 3y ago|0 comments
107.
Imperative, Declarative, Interrogative, and Exclamatory Interfaces(matt-rickard.com)
1 points by rckrd 3y ago|0 comments
108.
IaC: Strength or Weakness for Cloud Providers?(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
109.
On Unoptimizing(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
110.
Wrong Sequences for Startups(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
111.
Reasons to Do a Startup(matt-rickard.com)
2 points by rckrd 4y ago|1 comments
112.
The Missing Semester of CS(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
113.
Kubernetes as a Dev Tool(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
114.
Why DSLs Fail(matt-rickard.com)
13 points by rckrd 4y ago|17 comments
115.
Why Python Won't Be the Language of LLMs(matt-rickard.com)
6 points by rckrd 4y ago|3 comments
116.
GPT Lineage(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
117.
Docker Without Docker(matt-rickard.com)
2 points by rckrd 4y ago|0 comments
118.
If It Doesn’t Ship, It Doesn’t Exist(matt-rickard.com)
3 points by rckrd 4y ago|0 comments
119.
The Front End Bundling Cycle(matt-rickard.com)
1 points by rckrd 4y ago|0 comments
120.
The Front End Ouroboros(matt-rickard.com)
1 points by rckrd 4y ago|0 comments

More ›