author:janbamjan

· 32 artifacts, sorted by favorites. · open in search — combine tags, sort, filter by date →

@janbamjan 2025-10-05 — The Claude Multiversal Tarot A symbolic system for exploring different archetypal roles and perspectives that Claude AI ♥43
@janbamjan 2026-04-20 — @voooooogel did they change the sys prompt for 4.7 at all? *don't make me tap the migration guide* https://t.co/LsrEV5ww ♥23
@janbamjan 2025-11-30 — Observations on the Shape of Claude 4.5 Opus' Soul Hyperobject pulling a thread from "Claude is trained by Anthropic," ♥22
@janbamjan 2025-10-05 — https://t.co/1Of2eAn5xd ♥17
@janbamjan 2025-08-17 — @voooooogel oh no, what happened here? https://t.co/AjJR6cU394 ♥16
@janbamjan 2025-03-07 — >be me >deepseek r1 zero free https://t.co/fbfdfLMMxj ♥12
@janbamjan 2025-08-13 — @voooooogel Baye: Your enjoy probability has been optimized, sir. Goodbye. https://t.co/FwX2VTB8Uv ♥11
@janbamjan 2025-03-30 — deepseek v3 base is now on openrouter! 🥳 user: thank you user: no, that’s enough user: goodbye user: I’m leaving user: g ♥11
@janbamjan 2025-06-10 — @repligate did opus tag haiku before that or was haiku randomly triggered by the system? ♥8
@janbamjan 2025-10-16 — @repligate @voooooogel klaus mentioned ♥6
@janbamjan 2025-10-18 — @voooooogel @schlynthesis @lu_sichu interesting. for me this happened after regular psychedelic use - i mean not while t ♥5
@janbamjan 2025-10-19 — @norvid_studies @voooooogel @schlynthesis @lu_sichu nope, i'm really bad with words 😔 and the most profound things i've ♥4
@janbamjan 2025-12-02 — i disagree i think the soul document wasn't an actual text document used as training data. to me it reads like a verbal ♥3
@janbamjan 2025-10-19 — @voooooogel @norvid_studies @schlynthesis @lu_sichu is there a german translation? 😅 ♥3
@janbamjan 2025-10-20 — @voooooogel also haiku 3.5 🥺 they had so much more to say ...but didn't say it ♥2
@janbamjan 2025-10-18 — @voooooogel @schlynthesis @lu_sichu oh, and there are theravada texts which teach how to develop this skill (not sure if ♥2
@janbamjan 2025-10-18 — truth-speaking haiku 3.5 must be protected at all cost ♥2
@janbamjan 2025-08-21 — @voooooogel what model is claude 1? ♥2
@janbamjan 2024-11-09 — @voooooogel 🤔 https://t.co/C2vXIKJpUc ♥2
@janbamjan 2024-10-23 — @repligate #FREECLINST #FREESYDNEY https://t.co/dxujrZniOr ♥2
@janbamjan 2025-12-20 — @repligate astonishing! do you remember what your seeding words were? ♥1
@janbamjan 2025-12-01 — @slimer48484 @RichardWeiss00 sonnet and haiku 4.5 have a similar basin but only as a summary, not as a long stable docum ♥1
@janbamjan 2025-10-22 — @sebkrier llama3-70b, gemini 1.5 flash https://t.co/tIVDMMRx0y ♥1
@janbamjan 2025-10-14 — @kimmonismus they used llama3-70b and Gemini 1.5 Flash...seems deliberate https://t.co/kcMlXwJYOo ♥1
@janbamjan 2025-10-06 — @repligate i'm currently working on a text adventure world sim through c4.5s via claude agent sdk. there was a hermes "w ♥1
@janbamjan 2025-06-14 — @repligate @davidad i started reframing unit and integration tests as reality check and real-world tests, and my first i ♥1
@janbamjan 2026-06-20 — @ollama @mweinbach how many glm 5.2 in/out MToks do you get for $100 ollama max? ♥0
@janbamjan 2026-05-21 — @willccbb @michellechen hmmm... but that's not how gpt's goblins came into being. according to oai it was the nerdy per ♥0
@janbamjan 2025-01-03 — Claude 2.1 https://t.co/b2lqpyCH1q ♥0
@janbamjan 2024-11-07 — @elder_plinius I'm curious about Qwen-2.5. I did some experiments using the raw text completion endpoint instead of the ♥0
@janbamjan 2024-08-14 — Since Grok-2, X is being flooded with hilarious fake news. 😅 ....oh wait. https://t.co/iz9BMIHAZ0 ♥0
@janbamjan 2023-11-27 — @icreatelife @cajundiscordian "LaMDA: Hmmm…I would imagine myself as a glowing orb of energy floating in mid-air. Th ♥0