Christified

Everything is about Jesus

[Christified @mental_models_7]

@Pflxck @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon Model: gemma-4-E2B-it-Q4_K_M.gguf,
--temp 0
--seed 0
CPU bound
TOKENIZERS_PARALLELISM=false
Memoria: CPU-only

"Same user prompt, same thinking process, same answer, Twice"

Did i miss something?

[Christified @mental_models_7]

@22222222222223e @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon Yeah, it's true, what is the problem? If you feed a model with diferent prompts you will likely always have a different answer. Things like random seed also affect it.

[Christified @mental_models_7]

@22222222222223e @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon Someone told llms always gives different answers, I tried to explain the reason public llms are like that is intencional and can perfectly be otherwise having the same exact answer.

[Christified @mental_models_7]

@Sexinator1337 @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon Your comments only adds to mine

[Christified @mental_models_7]

@Pflxck @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon Now we are getting obtuse? Do you think if i even get rid of every volatile parameter (null or put it 0 like temperature) I won't get the exact same answer letter by letter?

[Christified @mental_models_7]

@Pflxck @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon LLMs are ultimately mathematics. In a complete controlled environment which you control hardware and software and metadata, yes they are deterministic, but in practice responses are quite affected by many volatile things.

[Christified @mental_models_7]

@22222222222223e @SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon When i said exact same data i also included seed

[Christified @mental_models_7]

@pseudo_maestro @megufl @DmitriyLeybel @Saanvi_dhillon Can you try with a "temporal session"? Same question? Temporal sessions are not influenced by previous user contexts.

[Christified @mental_models_7]

The notion of making a what if which you accept the all powerful - all knowing God exists and have the audacity to be against Him is the most retarded thing the atheist mind will bring you.

[Joakim 🌹🇳🇴🇪🇺 @joakial_]

If God were real, I’d despise him. Given that he is supposedly all-powerful, all-good, and all-knowing, I can’t reconcile the immense suffering in the world with the idea that such a being is worthy of worship or admiration.

[Christified @mental_models_7]

@SatikVFX_ @pseudo_maestro @DmitriyLeybel @Saanvi_dhillon I'm a developer, models are deterministic if you feed it with the exact same data always. What happend here is a combination between differents personalized context users collected through other sessions (which Grok has the ability) and possibly indeterministic web research.

[Christified @mental_models_7]

@pseudo_maestro @DmitriyLeybel @Saanvi_dhillon Just tried it. Sure you didn't fabricated it?

[Christified @mental_models_7]

@DaedricDrake Just by curiosity, how much time was alive inside?

[Christified @mental_models_7]

@FreedomNinja4 @akafaceUS Pretty much agree. Animals are animals. Just for personal matters I wouldn't eat cats or dogs in a normal situation, but if I'm literally starving I would even eat meat horse

[Christified @mental_models_7]

@Strepriptor @Bing57808343057 @TTExulansic What do you have to say to the ones who regret it? Come on, you can't have everyone happy. Cause' due we have results which not everyone was ok with changes for drugs and/or mutilations after they got it, you must accept is was "optional"

[Christified @mental_models_7]

@Strepriptor @Bing57808343057 @TTExulansic Your whole shit breaks when detransitioners that "regret" their decisions exists. You are not allowed to see and eat half the cake

[Christified @mental_models_7]

@Strepriptor @Bing57808343057 @TTExulansic You have it backwards. The fact they are choosing this is because they are not sane in the head.

[Christified @mental_models_7]

@FishAudio Fish

[Fish Audio @FishAudio]

Today we’ve raised $52M Seed and we are announcing the public launch of S2.1 Pro.

>It can clone a voice from 5 seconds of audio
>2x faster than Cartesia & 1/6th the cost of Eleven Labs
>most expressive model with word level control over emotion, intonation, pacing etc

We support frontier AI companies including HeyGen, LiveKit, Retell, Sanas, and OpenArt all run our model in production.

If you're a business and we can't cut your voice AI costs by 50%, we'll give you 1 year of Fish Audio for free.
Book a demo: https://t.co/vHkyZf9JoG

To celebrate our first birthday, we'll give you 1 month of S2.1 Pro for free. Like, retweet, and comment “Fish” to get it.

[Christified @mental_models_7]

@And13loq @Rx410 You won't be "living"

[Christified @mental_models_7]

You won't be "living"

[𝓐𝓷𝓭𝓲𝓮 🏳️‍⚧️ 🇵🇸🚩 @And13loq]

@Rx410 I'd prefer living among muslims than homphobic bigots