mmis1000

For my personal experience, antropic model have better user experience except for 4.7 and 4.8 though. 4.7 and 4.8 feels like expensive downgrade of 4.6 to me (I didn't know why these two should even exist)

However it's less willing to obey your instruction so it's less usable for general runtine flows.

Although the underlying model is unknown. If it expose input token count, the tokenizer may be probable though. Most tokenizer segemnts wildly different in CJK inputs. It can probably be used to fingerprint the tokenizer based on token count if it is using existing tokenizer.

The way jev actually internally work could be interesting though. I believe most llm are only tuned to return the first or second logits(or a few more) correctly as that is what the sampler would choose anyway. Do they alter existing model for better behavior across all options? Or they distilled one to have the proper behavior? We can only guess without the actual implementation.

You don't even bother text after the [a] at first place in this case

Your question is something like

anwser only a,b,c,d for following question a. b. c. d....

the model output possibility of next character a: 0.8 b: 0.7 c: 0.3 f: 0.2 d: 0.1

If the list contains option you did not provide. The model is confused anyway, it don't matter if you use grammer to filter out the bad option or not, the answer is screwed already.

If you really know what grammers did, grammer is a filter to mask out option llm provided but you don't like.

It does not change potential distribution in any means. It DROPS part of answer model returned directly.

The text generation model go wild because model relies on previous section it answered to continue later section. And because now it contain item model have no idea, it is completely screwed.

In the case you only require model to answer one of a,b,c,d and don't care about later segment at all. It don't really matter.

Fixed guard today is not very fixed. For ex, the safeguard qwen released is a full 4b llm model. It has no different to normal llm model arch except tuned for this specific purpose,

i mean webKit never handle viewport well when compare to firefox or chrome. it's willing to materialize a 10000*10000 texture when firefox and chrome will think that is too dumb and only load the visible part. It's not a good target if you want to compare windows and mac software quality.

I doubt the purpose is even age verification. They already can know I have a visa card which is credit card that can be only applied by an adult by looking at the BIN. Why they need my id card if the purpose is actually age verification?

Plus the routeros supports exporting all settings directly as text file. so you can really export the whole system, send it to llm and ask it to edit. install the new rule.

in worst case you just install the old rule back if it fuck up, no really damage can happen.

Every single asr model I tested so far did not support timestamps properly though. Some use external aligner to create timestamp, but the accuracy is still much inferior than whipser in case the audio is noisy.

Chrome is the only browser that preserve unused bit in value NaN through non JITed mode as far as I remember. And that bit become 0 when code get JITed.

browser report where your browser window (not position in tab) is even page itself never need to use it.

https://developer.mozilla.org/zh-CN/docs/Web/API/Window/scre...

If calling somebody with phone is still 'dialing' someone even there is nothing round on smart phone. Then why not?

Vue solve single source of truth much better than react though. The shape of derived data and source data (or even external data) are the same and interchangeable. So you can write an api that works with both without handling implementation quirks at all.

The words borrowed from someone else:

The react is more about view.

The vue is more about reactivity

> This started breaking down after ~30 files.

Codex's short context and todolist system combined somehow helps here though. Because of the frequent compact. The model was forced to recheck what todo list item has not done yet and what workflow skill it has to use. I used to left it for multi hour to do a big clean up and it finished without obvious issues.

Compatibility so foundation packages like torch onnx-runtime can run on AMD GPU without massive change in architecture. It's the biggest reason for those stuff that "only works on nvidia gpu". It's not faster if vulkan alternative exists, but at least it runs.

At least they finally do something this time. Now torch and whatever transformer stuff runs normally on windows/linux as long as you installed correct wheel from amd's own repository.

It's a huge step though.

I think the website should probably mention those installation preset in unsloth pyproject.toml though. The website instruct you to install dependencies separately. But it turns out there are dedicated preset that install specific rocm/cuda/xpu version in the project.

Uv helps you up though. Use a pyproject.toml and uv sync. Everything will be put into the venv only, nothing spread across the whole system.

The pyproject.toml can even handles build env for you, so you no longer need a setup.sh that installs 10 tool in specific order with specific flag to produce working environment. A single uv sync, and the job is done.

Plus the result is reproducible, so if this time uv sync work, then it also work next time.

Highly recommend if you are still on pip.

Note: Take a example that I used to install unsloth with rocm setup that based on unreleased git version dependencies and graphic card specific build flag, all of them can be handled with one command 'uv sync'. This will require a big pile of shell script if doing another way. https://github.com/unslothai/unsloth/issues/4280#issuecommen...

some come with only very short description but most part are only discoverable by 'man'.

and windows mostly use \? and also \h,

java user single - for long argument because it don't have short one.

I doubt it is ever close to reusable.

And even allowed position of parameters (or even meaning of arguments in case of ffmpeg) are program dependent.

Some allow anywhere as long as it is started with a dash, some only allow before first input

I think part of the problem is how these mcp service are designed. A lot of them just returns Mbs of text blob without filtering at all, and thus explodes the context.

And it's also affected by how model is trained. Gemini specifically like to read large amount of text data directly and explodes the context. But claude try to use tool for partial search or write a script to sample from a very large file. Gemini always fills the context way faster then claude when doing the same job.

But I guess in case of a bad designed mcp, there is no much model can do because the results are injected into context directly though (unless the runtime decided to redirect it to somewhere else)

Tell me how many ways that print help message for a command you have seen and say "reusable" again. Mcp is exactly exists to solve this. The rest is just json rpc with simple key value pairs.

You can probably let llm guess the help flag and try to parse help message. But the success rate is totally depends on model you are using.

If it explore all these cases after a few month and made the tool itself obsolete, that sounds like a total win to me?

However that don't happen unless firefox just stop developing though. New code comes with new bug, and there must be some people or some tool to find it out.

It's not really bad or not though. It's a more directed than the rest fuzzer. While being able to craft a payload that trigger flaw in deep flow path. It could also miss some obvious pattern that normal people don't think it will have problem (this is what most fuzzer currently tests)

> Only complaint is it sometimes decides to ignore half your prompt when instructions get long

This sounds like your context is too big and getting cut off.

Wondering how it decided to show the force exit program dialog. I used to use 8g macbook for development. But instead warning on serious memory exhaustion, it just decided to lag and suicide with everything freezed (including the restart button).

I think AMD just add support of rocm to rdna2 recently? I can run torch and aisudio with it just fine.

They also finally fix all ai related stuff building on windows, so you are no longer limited to linux for these.

from my personal experience, qwen 30b a3b understand command quiet well as long as the input is not big enough that ruin the attention (I feel the boundary is somewhere between 8000 or 12000?). But that isn't really bug of model itself though. A smaller model just have shorter memory, it's simply physical restriction.

I made a mixed extraction, cleaning, translation, formatting task on job that have average 6000 token input. And so far, only 30b a3b is smart enough not miss job detail (most of time)

I later refactor the task to multi pass using smaller model though. Make job simpler is still a better strategy to get clean output if you can change the pipeline.

windows does support [win] + [arrow key] though

This article always end up relevant once in a while.

Recently, I am trying to make llm to output specific format.

It turns out no matter how you wrote propmt and perform validate. It will never be as effective as just limit the output with proper bnf (via llama cpp grammar file).