Link the master Dataset Navigator

README.md

@@ -1,10 +1,149 @@
 ---
 title: Dataset Navigator
-emoji: 🌖
-colorFrom: gray
-colorTo: yellow
+emoji: 🗂️
+colorFrom: green
+colorTo: gray
 sdk: static
+app_file: index.html
 pinned: false
+license: cc-by-4.0
+short_description: Browse Daniel Rosehill's public datasets
+datasets:
+  - danielrosehill/Agent-Task-Break-Detector
+  - danielrosehill/AI-Interactions-Sentences
+  - danielrosehill/American-Centrism-Eval
+  - danielrosehill/Annotated-Voice-Notes
+  - danielrosehill/ASR-Paragraph-Breaks
+  - danielrosehill/ASR-WPM-And-Background-Noise-Eval
+  - danielrosehill/Audio-Understanding-Bitrate-Eval-0426
+  - danielrosehill/Audio-Understanding-Test-Set
+  - danielrosehill/Code-Gen-Agents-0925
+  - danielrosehill/Conspiracy-Endorsement-Eval
+  - danielrosehill/Conspiracy-Endorsement-Eval-Runs
+  - danielrosehill/Email-Key-Line-Bolding
+  - danielrosehill/English-Hebrew-Codeswitching-Sentences
+  - danielrosehill/General-English-Sentences
+  - danielrosehill/Geopol-Forecaster-Predictions
+  - danielrosehill/Geopolitical-System-Prompts
+  - danielrosehill/GHG-Emissions-Data
+  - danielrosehill/Global-Value-Factor-Database-Refactor-V2
+  - danielrosehill/Hebrew-Image-Eval-111225
+  - danielrosehill/Hebrew-Language-Signage
+  - danielrosehill/homegoods-taxonomy
+  - danielrosehill/Homophone-Context-Sentences
+  - danielrosehill/inventory-spec-lines
+  - danielrosehill/Iran-Israel-War-2026
+  - danielrosehill/ISO-3166-4217-Consolidated
+  - danielrosehill/Israel-Alerting-Zones
+  - danielrosehill/Israel-Election-2026-Candidate-Lists
+  - danielrosehill/Israel-Election-2026-Polls
+  - danielrosehill/Israel-Knesset-25-Members
+  - danielrosehill/Israel-Open-Data-Catalogue
+  - danielrosehill/Israel-Palestine-Framing-Eval
+  - danielrosehill/Israel-Palestine-Framing-Eval-Runs
+  - danielrosehill/Israel-Photos
+  - danielrosehill/Ivory-Catalog-020226
+  - danielrosehill/Jerusalem-Air-Quality-Shabbat
+  - danielrosehill/Jerusalem-Emergency-Shelters-0925
+  - danielrosehill/Jerusalem-High-Rise-Development
+  - danielrosehill/jerusalem-poster-detection
+  - danielrosehill/Jerusalem-Streetscapes
+  - danielrosehill/Linux-Admin-Sentences
+  - danielrosehill/Long-Prompt-Experiment
+  - danielrosehill/multimodal-ai-taxonomy
+  - danielrosehill/Noisy-Voice-Notes
+  - danielrosehill/NVR-Entity-Recognition-Experiment
+  - danielrosehill/Open-Router-API-Pricing-Analysis
+  - danielrosehill/OR-Batch-Inference-Models
+  - danielrosehill/Paragraph-Dictation-Passages
+  - danielrosehill/Parakeet-Dictation-Cleanup-Eval
+  - danielrosehill/pay-for-outcomes-instruments
+  - danielrosehill/Personal-Vocabulary-Sentences
+  - danielrosehill/Podcast-ASR-Evaluation
+  - danielrosehill/Prompt-Separation
+  - danielrosehill/Proofed-ASR-Cleanup-Training-Data
+  - danielrosehill/Propaganda-Bait-Eval
+  - danielrosehill/Propaganda-Bait-Eval-Runs
+  - danielrosehill/pseudo-personalised-outreach
+  - danielrosehill/Read-Aloud-Voice-Sentences
+  - danielrosehill/Sample-Voice-Context-Data
+  - danielrosehill/Shakespearean-Text-Transformation-Prompts
+  - danielrosehill/Single-Prompt-Book
+  - danielrosehill/Small-STT-Eval-Audio-Dataset
+  - danielrosehill/Smart-Home-Sentences
+  - danielrosehill/Speech-To-Text-System-Prompts-2
+  - danielrosehill/storage-container-dimensions
+  - danielrosehill/STT-Fine-Tune-Eval-101225
+  - danielrosehill/STT-Voice-Notes-Evals
+  - danielrosehill/System-Prompt-Library
+  - danielrosehill/Tech-Graphics-Benchmark
+  - danielrosehill/Tech-Graphics-Benchmark-Ratings
+  - danielrosehill/Tech-Graphics-Benchmark-Runs
+  - danielrosehill/Tech-Sentences
+  - danielrosehill/Tech-Sentences-For-ASR-Training
+  - danielrosehill/Tel-Aviv-Pics
+  - danielrosehill/term-context-sentences
+  - danielrosehill/Text-To-Image-Test-Prompts
+  - danielrosehill/Text-Transformation-Prompts-300525
+  - danielrosehill/The-PC-Bench
+  - danielrosehill/Transcription-Cleanup-Trainer
+  - danielrosehill/US-Partisan-Framing-Eval
+  - danielrosehill/Voice-Command-Sentences
+  - danielrosehill/Voice-Note-Audio
+  - danielrosehill/Voice-Notes
+  - danielrosehill/Wake-Words
+  - danielrosehill/Whisper-Fine-Tune-One-Shot-Eval
+  - danielrosehill/Whiteboards
+  - danielrosehill/Writing-System-Prompts
+  - danielrosehill/Zapier-Integrations-260825
+  - Hebrish/Acronym-Sentences
+  - Hebrish/AI-Interactions-Sentence-Bank
+  - Hebrish/AI-Tech-Vocabulary-Sentence-Bank
+  - Hebrish/ASR-Error-List
+  - Hebrish/ASR-Error-Prediction
+  - Hebrish/Brand-Name-Sentences
+  - Hebrish/Community-Voice-Contributions
+  - Hebrish/Currency-Sentences
+  - Hebrish/Disfluent-Read-Sentences
+  - Hebrish/English-Hebrew-Codeswitching-ASR-Benchmark
+  - Hebrish/English-Hebrew-Mixed-Sentences
+  - Hebrish/Example-Sentence-Bank
+  - Hebrish/Expanded-Acronym-Sentences
+  - Hebrish/Hard-ASR-Sentences
+  - Hebrish/hebrew-latin-code-switching
+  - Hebrish/Hebrew-Term-Pronunciations-IPA
+  - Hebrish/Hebrish-TTS-Voice-Recordings
+  - Hebrish/Hebrish-Validation
+  - Hebrish/Homophone-Context-Sentence-Bank
+  - Hebrish/Irish-English-Sentences
+  - Hebrish/Israeli-Name-Sentences
+  - Hebrish/Israeli-Street-Names
+  - Hebrish/Jewish-Religious-Terminology-Sentences
+  - Hebrish/Judaica-Sentences
+  - Hebrish/Linux-Admin-Sentence-Bank
+  - Hebrish/Noisy-Condition-Sentences
+  - Hebrish/October-7-Memorial-Sentence-Bank
+  - Hebrish/Personal-Vocabulary-Sentence-Bank
+  - Hebrish/Pronunciation-Variants
+  - Hebrish/Repeated-Word-Sentences
+  - Hebrish/Room-Tone-Annotations
+  - Hebrish/Self-Correction-Sentences
+  - Hebrish/Singing-Sentences
+  - Hebrish/Smart-Home-Sentence-Bank
+  - Hebrish/Spoken-Aside-Sentences
+  - Hebrish/Synthetic-Codeswitching-Voices-Pilot
+  - Hebrish/Weird-Voice-Sentences
+  - Hebrish/Whispered-Sentences
+  - Hebrish/Yiddish-Sentences
+  - My-Weird-Prompts/Corn-The-Sloth
+  - My-Weird-Prompts/Corn-Training-Set
+  - My-Weird-Prompts/episodes
+  - My-Weird-Prompts/episodes-eval
+  - My-Weird-Prompts/Eval-Prompts
+  - My-Weird-Prompts/Model-Chronicle
+  - My-Weird-Prompts/mwp-censorship-results
+  - My-Weird-Prompts/MWP-Script-Results
+  - My-Weird-Prompts/transcripts
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Master navigator for Daniel Rosehill's public datasets, on the `danielrosehill` account and the Hebrish and My-Weird-Prompts organisations. It opens on the specialist navigators, and topic tabs cover the rest.

app.js

@@ -0,0 +1,48 @@
+const HF = "https://huggingface.co/";
+const esc = (s) => String(s ?? "").replace(/[&<>"]/g, (c) => ({ "&": "&amp;", "<": "&lt;", ">": "&gt;", '"': "&quot;" }[c]));
+const link = (path, text) => `<a href="${esc(HF + path)}" target="_blank" rel="noopener">${esc(text)}</a>`;
+const size = (b) => (b >= 1e9 ? (b / 1e9).toFixed(1) + " GB" : b >= 1e6 ? Math.round(b / 1e6) + " MB" : Math.max(1, Math.round(b / 1e3)) + " KB");
+const logo = (o) => (o === "danielrosehill" ? '<img src="daniel.webp" alt="">' : o === "Hebrish" ? '<img src="hebrish.webp" alt="">' : "");
+
+function card(d) {
+  const facts = [["Size", size(d.bytes)]];
+  if (d.rows >= 10) facts.push(["Rows", d.rows.toLocaleString("en-US")]);
+  if (d.license) facts.push(["License", esc(d.license)]);
+  return `<article class="card"><div class="who">${logo(d.owner)}<span>${esc(d.owner)} · ${esc(d.type)}</span></div>
+    <h3>${link("datasets/" + d.id, d.name)}</h3><p>${esc(d.summary)}</p>
+    ${d.gated ? '<div class="meta"><span class="badge">Access request</span></div>' : ""}
+    <dl class="facts">${facts.map(([k, v]) => `<dt>${k}</dt><dd>${v}</dd>`).join("")}</dl></article>`;
+}
+const navCard = (n) => `<article class="card"><div class="who"><span>Navigator</span></div><h3>${link("spaces/" + n.space, n.name)}</h3><p>${esc(n.summary)}</p>
+  <div class="count">${n.count} <span>dataset${n.count === 1 ? "" : "s"}</span></div></article>`;
+const topicCard = (g) => `<article class="card"><div class="who"><span>Topic</span></div><h3><a href="#${g.slug}">${esc(g.name)}</a></h3>
+  <p>${esc(g.datasets.slice(0, 3).map((d) => d.name).join(", "))} and more.</p><div class="count">${g.datasets.length} <span>datasets</span></div></article>`;
+
+(async () => {
+  const data = await (await fetch("data.json")).json();
+  const q = document.getElementById("q"), out = document.getElementById("list"), tabs = document.getElementById("tabs");
+  const navName = Object.fromEntries(data.navigators.map((n) => [n.id, n.name]));
+  document.getElementById("built").textContent = data.built;
+  document.getElementById("count").textContent = `${data.total} datasets.`;
+  const grid = (html) => `<div class="grid">${html}</div>`;
+  const draw = () => {
+    const slug = location.hash.slice(1), current = data.groups.find((g) => g.slug === slug);
+    const term = q.value.trim().toLowerCase();
+    tabs.innerHTML = [`<a href="#"${!term && !current ? ' aria-current="page"' : ""}>Navigators</a>`,
+      ...data.groups.map((g) => `<a href="#${g.slug}"${!term && g === current ? ' aria-current="page"' : ""}>${esc(g.tab)} (${g.datasets.length})</a>`)].join("");
+    if (!term) {
+      out.innerHTML = current ? `<h2>${esc(current.name)}</h2>` + grid(current.datasets.map(card).join(""))
+        : "<h2>Specialist navigators</h2>" + grid(data.navigators.map(navCard).join("")) + "<h2>By topic</h2>" + grid(data.groups.map(topicCard).join(""));
+      return;
+    }
+    const has = (...parts) => parts.join(" ").toLowerCase().includes(term);
+    const groups = data.groups.map((g) => [g, g.datasets.filter((d) => has(d.id, d.name, d.type, d.summary))]).filter(([, h]) => h.length);
+    const inside = data.inside.filter((d) => has(d.id, d.name, navName[d.nav]));
+    out.innerHTML = groups.map(([g, h]) => `<h2>${esc(g.name)}</h2>` + grid(h.map(card).join(""))).join("")
+      + (inside.length ? "<h2>Inside the navigators</h2>" + grid(inside.map((d) => `<article class="card"><div class="who"><span>${esc(d.id.split("/")[0])} · ${esc(navName[d.nav])}</span></div><h3>${link("datasets/" + d.id, d.name)}</h3></article>`).join("")) : "");
+    if (!out.innerHTML) out.innerHTML = '<p class="empty">No datasets match.</p>';
+  };
+  q.addEventListener("input", draw);
+  addEventListener("hashchange", () => { q.value = ""; draw(); });
+  draw();
+})();

data.json

@@ -0,0 +1,1217 @@
+{
+ "built": "2026-11-07",
+ "total": 135,
+ "navigators": [
+  {
+   "id": "audio",
+   "name": "Audio Dataset Navigator",
+   "space": "danielrosehill/Audio-Dataset-Navigator",
+   "summary": "Read-sentence voice recordings, for training and testing speech recognition.",
+   "count": 37
+  },
+  {
+   "id": "sentence",
+   "name": "Sentence Dataset Navigator",
+   "space": "danielrosehill/Sentence-Dataset-Navigator",
+   "summary": "Text-only sentence banks for reading aloud, and third-party candidates.",
+   "count": 8
+  },
+  {
+   "id": "benchmarks",
+   "name": "My Benchmarks",
+   "space": "danielrosehill/My-Benchmarks",
+   "summary": "Test sets, prompts and run results, grouped by benchmark.",
+   "count": 23
+  },
+  {
+   "id": "mwp",
+   "name": "My Weird Prompts Dataset Navigator",
+   "space": "danielrosehill/mwp-dataset-navigator",
+   "summary": "Field reference for the podcast's two open datasets: every episode and every speaker turn.",
+   "count": 2
+  }
+ ],
+ "groups": [
+  {
+   "slug": "speech",
+   "tab": "Speech",
+   "name": "Speech and voice data",
+   "datasets": [
+    {
+     "id": "danielrosehill/Annotated-Voice-Notes",
+     "type": "Voice notes",
+     "owner": "danielrosehill",
+     "name": "Annotated Voice Notes",
+     "summary": "A small, hand-picked set of real voice notes. It is kept small on purpose.",
+     "gated": false,
+     "rows": 1,
+     "bytes": 21457417,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-03"
+    },
+    {
+     "id": "Hebrish/ASR-Error-List",
+     "type": "ASR errors",
+     "owner": "Hebrish",
+     "name": "ASR Error List",
+     "summary": "The words the Hebrish speech-to-text fine-tunes still get wrong in real dictation, what they were heard as, and how far targeted retraining on each one has got.",
+     "gated": true,
+     "rows": 2316,
+     "bytes": 816145,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-04"
+    },
+    {
+     "id": "Hebrish/Community-Voice-Contributions",
+     "type": "Recordings",
+     "owner": "Hebrish",
+     "name": "Community Voice Contributions",
+     "summary": "A place for anyone to contribute recordings of their own voice to the Hebrish speech models: English sentences with everyday Hebrew words written in Latin letters (\"the arnona bill\", \"a tor at the…",
+     "gated": false,
+     "rows": 1,
+     "bytes": 110593,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-02"
+    },
+    {
+     "id": "Hebrish/Room-Tone-Annotations",
+     "type": "Noise",
+     "owner": "Hebrish",
+     "name": "Room Tone Annotations",
+     "summary": "Room tone: recordings of silence in a labelled room, made with the same phone, microphone and settings as the Hebrish read-speech recordings. Each recording is tagged with the room, the conditions at…",
+     "gated": false,
+     "rows": 2,
+     "bytes": 4848280,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-22"
+    },
+    {
+     "id": "danielrosehill/Sample-Voice-Context-Data",
+     "type": "Voice context",
+     "owner": "danielrosehill",
+     "name": "Sample Voice Context Data",
+     "summary": "A small synthetic dataset containing LLM-generated context information simulating a job seeker narrating their career trajectory.",
+     "gated": false,
+     "rows": 18,
+     "bytes": 15809025,
+     "license": "mit",
+     "modified": "2026-10-30"
+    },
+    {
+     "id": "Hebrish/Synthetic-Codeswitching-Voices-Pilot",
+     "type": "Synthetic speech",
+     "owner": "Hebrish",
+     "name": "Synthetic Codeswitching Voices Pilot",
+     "summary": "100 synthetic clips, 2026-10-28: English sentences with one Hebrew term, spoken by Kokoro-82M voices (af_bella, af_heart, am_fenrir, am_michael, bf_emma, bm_george; US and UK, female and male), with…",
+     "gated": false,
+     "rows": 100,
+     "bytes": 17921280,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-28"
+    },
+    {
+     "id": "danielrosehill/Voice-Note-Audio",
+     "type": "Voice notes",
+     "owner": "danielrosehill",
+     "name": "Voice Note Audio",
+     "summary": "This dataset contains real-world voice recordings with transcripts and comprehensive annotations.",
+     "gated": false,
+     "rows": 101,
+     "bytes": 287702581,
+     "license": null,
+     "modified": "2026-10-30"
+    },
+    {
+     "id": "danielrosehill/Voice-Notes",
+     "type": "Voice notes",
+     "owner": "danielrosehill",
+     "name": "Voice Notes",
+     "summary": "Real voice-note transcripts from one person's Voicenotes export (April 2026), filtered for privacy, for evaluating Voice-Note-Intent-Classifier.",
+     "gated": false,
+     "rows": 65,
+     "bytes": 62057,
+     "license": "other",
+     "modified": "2026-11-03"
+    },
+    {
+     "id": "danielrosehill/Wake-Words",
+     "type": "Wake words",
+     "owner": "danielrosehill",
+     "name": "Wake Words",
+     "summary": "Real recordings of Jewish and Israeli first-name wake words, for training local wake word models (openWakeWord, microWakeWord) to use with Home Assistant Assist and on the desktop, in place of \"Hey…",
+     "gated": false,
+     "rows": 448,
+     "bytes": 122016836,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-30"
+    }
+   ]
+  },
+  {
+   "slug": "training",
+   "tab": "Training data",
+   "name": "Training data for small models",
+   "datasets": [
+    {
+     "id": "danielrosehill/Agent-Task-Break-Detector",
+     "type": "Classifier data",
+     "owner": "danielrosehill",
+     "name": "Agent Task Break Detector",
+     "summary": "Synthetic training data for a small classifier that judges whether a new user turn in an AI agent session starts a new task.",
+     "gated": false,
+     "rows": 8193,
+     "bytes": 18002594,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-24"
+    },
+    {
+     "id": "Hebrish/ASR-Error-Prediction",
+     "type": "ASR errors",
+     "owner": "Hebrish",
+     "name": "ASR Error Prediction",
+     "summary": "Training data for a model that predicts which words a speech recognition model will get wrong for one speaker, by extrapolating from the words it has already been confirmed to get wrong. Built for…",
+     "gated": false,
+     "rows": 9685,
+     "bytes": 2422400,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-25"
+    },
+    {
+     "id": "danielrosehill/ASR-Paragraph-Breaks",
+     "type": "Text clean-up",
+     "owner": "danielrosehill",
+     "name": "ASR Paragraph Breaks",
+     "summary": "Hand-annotated paragraph breaks for flat speech-to-text output. ASR engines emit punctuated, cased text as one block.",
+     "gated": false,
+     "rows": 1535,
+     "bytes": 8129048,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-25"
+    },
+    {
+     "id": "danielrosehill/Email-Key-Line-Bolding",
+     "type": "Text clean-up",
+     "owner": "danielrosehill",
+     "name": "Email Key Line Bolding",
+     "summary": "Hand annotations of which words in a dictated email should be bold: any run of whole words within a paragraph, from a phrase to several sentences. The task for the model trained on it: given an…",
+     "gated": false,
+     "rows": 342,
+     "bytes": 4361362,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-21"
+    },
+    {
+     "id": "Hebrish/Hebrew-Term-Pronunciations-IPA",
+     "type": "Pronunciation",
+     "owner": "Hebrish",
+     "name": "Hebrew Term Pronunciations IPA",
+     "summary": "How Hebrew words are pronounced inside English speech, written in IPA, measured from real recordings of one Anglo-Israeli speaker (Daniel Rosehill). Pilot of 20 terms, 2026-10-28.",
+     "gated": true,
+     "rows": 170,
+     "bytes": 70247,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-28"
+    },
+    {
+     "id": "Hebrish/hebrew-latin-code-switching",
+     "type": "Token labels",
+     "owner": "Hebrish",
+     "name": "Hebrew-Latin Code-Switching",
+     "summary": "English sentences containing Hebrew words written in Latin characters, with token-level BIO spans marking them.",
+     "gated": true,
+     "rows": 4737,
+     "bytes": 2313938,
+     "license": "mit",
+     "modified": "2026-10-30"
+    },
+    {
+     "id": "danielrosehill/inventory-spec-lines",
+     "type": "Token labels",
+     "owner": "danielrosehill",
+     "name": "Inventory Spec Lines",
+     "summary": "14,426 product-specification lines from a real household inventory (cables, adapters, tools, electronics, household goods), each labelled with spans for a fixed set of 25 attributes (vocabulary v2)…",
+     "gated": false,
+     "rows": 14426,
+     "bytes": 2444960,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-15"
+    },
+    {
+     "id": "danielrosehill/jerusalem-poster-detection",
+     "type": "Object detection",
+     "owner": "danielrosehill",
+     "name": "Jerusalem Poster Detection",
+     "summary": "Geotagged street photographs of a messianic postering campaign in central Jerusalem, with bounding boxes around every campaign poster in frame. Each image carries the GPS coordinates it was shot at…",
+     "gated": false,
+     "rows": 13,
+     "bytes": 11412223,
+     "license": "cc-by-4.0",
+     "modified": "2026-09-01"
+    },
+    {
+     "id": "danielrosehill/NVR-Entity-Recognition-Experiment",
+     "type": "Entity recognition",
+     "owner": "danielrosehill",
+     "name": "NVR Entity Recognition Experiment",
+     "summary": "This repository contains a training dataset designed for entity recognition in Network Video Recorder (NVR) applications, specifically focused on newborn safety monitoring. The dataset uses a stuffed…",
+     "gated": false,
+     "rows": 196,
+     "bytes": 196699027,
+     "license": "mit",
+     "modified": "2025-08-21"
+    },
+    {
+     "id": "danielrosehill/Prompt-Separation",
+     "type": "Text structure",
+     "owner": "danielrosehill",
+     "name": "Prompt Separation",
+     "summary": "Voice-typed podcast prompt transcripts decomposed into structured fields: discrete prompts (asks), a list of context chunks, and free-form host_notes.",
+     "gated": false,
+     "rows": 2318,
+     "bytes": 4961708,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-19"
+    },
+    {
+     "id": "danielrosehill/pseudo-personalised-outreach",
+     "type": "Classifier data",
+     "owner": "danielrosehill",
+     "name": "Pseudo-Personalised Outreach",
+     "summary": "Real unsolicited emails sent to one person, labelled genuine or pseudo_personalised, with every sender identity replaced by a plausible fake.",
+     "gated": false,
+     "rows": 18,
+     "bytes": 64651,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-20"
+    },
+    {
+     "id": "danielrosehill/term-context-sentences",
+     "type": "Sentence generation",
+     "owner": "danielrosehill",
+     "name": "Term Context Sentences",
+     "summary": "Training data for Contextual Sentence Creator: given a term, a one-line context and a count, write natural English sentences.",
+     "gated": false,
+     "rows": 21815,
+     "bytes": 20097789,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-03"
+    },
+    {
+     "id": "danielrosehill/Transcription-Cleanup-Trainer",
+     "type": "Text clean-up",
+     "owner": "danielrosehill",
+     "name": "Transcription Cleanup Trainer",
+     "summary": "A curated dataset for training speech-to-text cleanup models to achieve optimal transcript refinement.",
+     "gated": false,
+     "rows": 1,
+     "bytes": 3089244,
+     "license": null,
+     "modified": "2026-10-30"
+    }
+   ]
+  },
+  {
+   "slug": "israel",
+   "tab": "Israel",
+   "name": "Israel and geopolitics",
+   "datasets": [
+    {
+     "id": "danielrosehill/Geopol-Forecaster-Predictions",
+     "type": "Forecasting",
+     "owner": "danielrosehill",
+     "name": "Geopol Forecaster Predictions",
+     "summary": "Structured predictions extracted from multi-agent geopolitical wargaming simulations run by Geopol Forecaster.",
+     "gated": false,
+     "rows": 0,
+     "bytes": 59107,
+     "license": "apache-2.0",
+     "modified": "2026-04-13"
+    },
+    {
+     "id": "danielrosehill/Iran-Israel-War-2026",
+     "type": "OSINT",
+     "owner": "danielrosehill",
+     "name": "Iran Israel War 2026",
+     "summary": "Open-source intelligence dataset tracking Iranian missile and drone attack waves against Israel and US/coalition targets across four operations in the \"True Promise\" series (2024–2026).",
+     "gated": false,
+     "rows": 0,
+     "bytes": 665341,
+     "license": "cc-by-4.0",
+     "modified": "2026-03-18"
+    },
+    {
+     "id": "danielrosehill/Israel-Alerting-Zones",
+     "type": "Emergency",
+     "owner": "danielrosehill",
+     "name": "Israel Alerting Zones",
+     "summary": "This repository contains a comprehensive list of emergency alerting zones used in Israel by the Home Front Command (Pikud HaOref), compiled on May 9th, 2025.",
+     "gated": false,
+     "rows": 0,
+     "bytes": 66492,
+     "license": "apache-2.0",
+     "modified": "2025-05-09"
+    },
+    {
+     "id": "danielrosehill/Israel-Election-2026-Candidate-Lists",
+     "type": "Elections",
+     "owner": "danielrosehill",
+     "name": "Israel Election 2026 Candidate Lists",
+     "summary": "This dataset contains every candidate list for the election to the 26th Knesset, held on 27 October 2026. Each candidate appears with their list and position, and every row also names the list's…",
+     "gated": false,
+     "rows": 1399,
+     "bytes": 419489,
+     "license": "cc-by-sa-4.0",
+     "modified": "2026-10-22"
+    },
+    {
+     "id": "danielrosehill/Israel-Election-2026-Polls",
+     "type": "Elections",
+     "owner": "danielrosehill",
+     "name": "Israel Election 2026 Polls",
+     "summary": "This dataset contains every published seat-projection poll for the 27 October 2026 Knesset election, from 23 December 2022 to 22 October 2026. A publication ban starts at 23:59 on 23 October, so no…",
+     "gated": false,
+     "rows": 16517,
+     "bytes": 3354104,
+     "license": "cc-by-sa-4.0",
+     "modified": "2026-10-22"
+    },
+    {
+     "id": "danielrosehill/Israel-Knesset-25-Members",
+     "type": "Politics",
+     "owner": "danielrosehill",
+     "name": "Israel Knesset 25 Members",
+     "summary": "This dataset covers the outgoing 25th Knesset, elected on 1 November 2022, at the end of its term ahead of the 27 October 2026 election. It records everyone who held a seat and how the factions stood…",
+     "gated": false,
+     "rows": 168,
+     "bytes": 50299,
+     "license": "cc-by-sa-4.0",
+     "modified": "2026-10-22"
+    },
+    {
+     "id": "danielrosehill/Israel-Open-Data-Catalogue",
+     "type": "Open data",
+     "owner": "danielrosehill",
+     "name": "Israel Open Data Catalogue",
+     "summary": "A bilingual (Hebrew/English) snapshot of the entire dataset catalogue published on data.gov.il, Israel's official open-data portal.",
+     "gated": false,
+     "rows": 0,
+     "bytes": 23574528,
+     "license": "cc-by-4.0",
+     "modified": "2026-05-06"
+    },
+    {
+     "id": "Hebrish/Israeli-Street-Names",
+     "type": "Reference",
+     "owner": "Hebrish",
+     "name": "Israeli Street Names",
+     "summary": "Every street in Israel's national street list (63662 rows, 1316 localities, 31743 distinct names), with an English name from OpenStreetMap where one exists (9680 names, 35507 rows). Part of Hebrish…",
+     "gated": true,
+     "rows": 63662,
+     "bytes": 12175938,
+     "license": "other",
+     "modified": "2026-11-02"
+    },
+    {
+     "id": "danielrosehill/Jerusalem-Air-Quality-Shabbat",
+     "type": "Environment",
+     "owner": "danielrosehill",
+     "name": "Jerusalem Air Quality Shabbat",
+     "summary": "Twelve months of 5-minute-resolution air-quality readings from 12 monitoring stations across Jerusalem, formatted for analysing the weekday vs Friday vs Shabbat (Saturday) pattern in urban pollution.",
+     "gated": false,
+     "rows": 7949506,
+     "bytes": 64055809,
+     "license": "cc-by-4.0",
+     "modified": "2026-04-28"
+    },
+    {
+     "id": "danielrosehill/Jerusalem-Emergency-Shelters-0925",
+     "type": "Emergency",
+     "owner": "danielrosehill",
+     "name": "Jerusalem Emergency Shelters (September 2025)",
+     "summary": "This repository contains updated location data for public shelters in the Jerusalem area, populated on September 19th, 2025. The data has been processed and enhanced to support individual…",
+     "gated": false,
+     "rows": 149,
+     "bytes": 187134,
+     "license": null,
+     "modified": "2025-09-19"
+    }
+   ]
+  },
+  {
+   "slug": "images",
+   "tab": "Images",
+   "name": "Photo sets",
+   "datasets": [
+    {
+     "id": "danielrosehill/Hebrew-Language-Signage",
+     "type": "Photos",
+     "owner": "danielrosehill",
+     "name": "Hebrew Language Signage",
+     "summary": "This dataset contains photographs of Hebrew language text in everyday contexts throughout Israel, with a particular focus on signage displays including street signs, commercial signage, and public…",
+     "gated": false,
+     "rows": 68,
+     "bytes": 61421863,
+     "license": null,
+     "modified": "2025-11-06"
+    },
+    {
+     "id": "danielrosehill/Israel-Photos",
+     "type": "Photos",
+     "owner": "danielrosehill",
+     "name": "Israel Photos",
+     "summary": "A collection of 369 photographs captured across Israel between 2024 and 2025, with LLM-generated captions and location annotations. The images are sourced from the photographer's Pexels gallery.",
+     "gated": false,
+     "rows": 374,
+     "bytes": 513057983,
+     "license": "cc-by-4.0",
+     "modified": "2025-12-03"
+    },
+    {
+     "id": "danielrosehill/Jerusalem-High-Rise-Development",
+     "type": "Photos",
+     "owner": "danielrosehill",
+     "name": "Jerusalem High Rise Development",
+     "summary": "This dataset contains 56 photographs documenting high-rise buildings and urban development in Jerusalem, Israel. The images capture the architectural evolution of Jerusalem's modern skyline…",
+     "gated": false,
+     "rows": 56,
+     "bytes": 72331945,
+     "license": null,
+     "modified": "2025-11-06"
+    },
+    {
+     "id": "danielrosehill/Jerusalem-Streetscapes",
+     "type": "Photos",
+     "owner": "danielrosehill",
+     "name": "Jerusalem Streetscapes",
+     "summary": "A small image dataset containing photos of the rapidly changing urban landscape of Jerusalem, Israel, captured by day and by night.",
+     "gated": false,
+     "rows": 413,
+     "bytes": 436129089,
+     "license": null,
+     "modified": "2025-11-06"
+    },
+    {
+     "id": "danielrosehill/Tel-Aviv-Pics",
+     "type": "Photos",
+     "owner": "danielrosehill",
+     "name": "Tel Aviv Pics",
+     "summary": "This dataset contains 53 high-quality photographs of Tel Aviv's urban environment, captured to serve as reference material for game development, 3D world creation, and digital environment design.",
+     "gated": false,
+     "rows": 0,
+     "bytes": 180295574,
+     "license": "cc-by-4.0",
+     "modified": "2025-11-06"
+    },
+    {
+     "id": "danielrosehill/Whiteboards",
+     "type": "Photos",
+     "owner": "danielrosehill",
+     "name": "Whiteboards",
+     "summary": "A small, single-author whiteboard corpus for evaluating vision-language models on handwritten OCR accuracy and for studying pseudotext hallucination — the failure mode where a VLM invents…",
+     "gated": false,
+     "rows": 21,
+     "bytes": 3538210,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-03"
+    }
+   ]
+  },
+  {
+   "slug": "reference",
+   "tab": "Reference",
+   "name": "Reference data and taxonomies",
+   "datasets": [
+    {
+     "id": "danielrosehill/GHG-Emissions-Data",
+     "type": "Sustainability",
+     "owner": "danielrosehill",
+     "name": "GHG Emissions Data",
+     "summary": "This repository contains a comprehensive pipeline for processing and analyzing greenhouse gas (GHG) emissions data. The pipeline integrates datasets from multiple sources, including Climate TRACE and…",
+     "gated": false,
+     "rows": 78,
+     "bytes": 421897,
+     "license": "cc-by-4.0",
+     "modified": "2024-12-20"
+    },
+    {
+     "id": "danielrosehill/Global-Value-Factor-Database-Refactor-V2",
+     "type": "Sustainability",
+     "owner": "danielrosehill",
+     "name": "Global Value Factor Database (Refactor V2)",
+     "summary": "A refactor of the Global Value Factor Database from the International Foundation for Valuing Impacts.",
+     "gated": false,
+     "rows": 0,
+     "bytes": 180428515,
+     "license": "apache-2.0",
+     "modified": "2025-09-02"
+    },
+    {
+     "id": "danielrosehill/homegoods-taxonomy",
+     "type": "Taxonomy",
+     "owner": "danielrosehill",
+     "name": "HomeGoods Taxonomy",
+     "summary": "A hierarchical classification of the physical goods in an ordinary home or small business, with a physical handling attribute layer attached to every class.",
+     "gated": false,
+     "rows": 367,
+     "bytes": 847069,
+     "license": "mit",
+     "modified": "2026-08-03"
+    },
+    {
+     "id": "danielrosehill/ISO-3166-4217-Consolidated",
+     "type": "Lookup",
+     "owner": "danielrosehill",
+     "name": "ISO 3166 and 4217 Consolidated",
+     "summary": "This dataset contains a mapping between ISO-3166 (countries) and ISO-4217 (currencies).",
+     "gated": false,
+     "rows": 0,
+     "bytes": 313280,
+     "license": "apache-2.0",
+     "modified": "2025-09-03"
+    },
+    {
+     "id": "danielrosehill/Ivory-Catalog-020226",
+     "type": "Price snapshot",
+     "owner": "danielrosehill",
+     "name": "Ivory Catalog (February 2026)",
+     "summary": "A snapshot of computer hardware pricing from Ivory.co.il, one of Israel's largest electronics retailers, captured on February 2, 2026.",
+     "gated": false,
+     "rows": 9,
+     "bytes": 556453,
+     "license": "cc-by-4.0",
+     "modified": "2026-02-02"
+    },
+    {
+     "id": "danielrosehill/multimodal-ai-taxonomy",
+     "type": "Taxonomy",
+     "owner": "danielrosehill",
+     "name": "Multimodal AI Taxonomy",
+     "summary": "A comprehensive, structured taxonomy for mapping multimodal AI model capabilities across input and output modalities.",
+     "gated": false,
+     "rows": 85,
+     "bytes": 211526,
+     "license": "cc0-1.0",
+     "modified": "2025-10-22"
+    },
+    {
+     "id": "danielrosehill/Open-Router-API-Pricing-Analysis",
+     "type": "AI pricing",
+     "owner": "danielrosehill",
+     "name": "OpenRouter API Pricing Analysis",
+     "summary": "This dataset provides a point-in-time capture of pricing and parameters for LLMs available through the OpenRouter API for inference.",
+     "gated": false,
+     "rows": 2380,
+     "bytes": 1321036,
+     "license": "mit",
+     "modified": "2025-11-10"
+    },
+    {
+     "id": "danielrosehill/OR-Batch-Inference-Models",
+     "type": "AI pricing",
+     "owner": "danielrosehill",
+     "name": "OpenRouter Batch Inference Models",
+     "summary": "Every model with a :batch id on OpenRouter on 2026-10-31 (71 models), with batch and sync prices, the providers that serve batch, and a price-tier label.",
+     "gated": false,
+     "rows": 71,
+     "bytes": 38500,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-31"
+    },
+    {
+     "id": "danielrosehill/pay-for-outcomes-instruments",
+     "type": "Impact finance",
+     "owner": "danielrosehill",
+     "name": "Pay For Outcomes Instruments",
+     "summary": "This repository contains a curated, redacted, and standardized data set based on the Government Outcome Labs project at Oxford University (UK). It is the leading international data resource tracking…",
+     "gated": false,
+     "rows": 0,
+     "bytes": 3779175,
+     "license": "apache-2.0",
+     "modified": "2025-05-21"
+    },
+    {
+     "id": "danielrosehill/storage-container-dimensions",
+     "type": "Lookup",
+     "owner": "danielrosehill",
+     "name": "Storage Container Dimensions",
+     "summary": "Planning-grade volumetric reference for the plastic containers that European and North American warehouses actually run on: Euroboxes (Euro stacking containers), attached-lid containers (ALCs), and…",
+     "gated": false,
+     "rows": 169,
+     "bytes": 750079,
+     "license": "cc-by-4.0",
+     "modified": "2026-07-30"
+    },
+    {
+     "id": "danielrosehill/Zapier-Integrations-260825",
+     "type": "Snapshot",
+     "owner": "danielrosehill",
+     "name": "Zapier Integrations (August 2025)",
+     "summary": "Zapier apps and integrations, extracted on 26 August 2025.",
+     "gated": false,
+     "rows": 8643,
+     "bytes": 10991528,
+     "license": "apache-2.0",
+     "modified": "2025-08-26"
+    }
+   ]
+  },
+  {
+   "slug": "prompts",
+   "tab": "Prompts",
+   "name": "Prompt libraries and experiments",
+   "datasets": [
+    {
+     "id": "danielrosehill/Code-Gen-Agents-0925",
+     "type": "Agent configs",
+     "owner": "danielrosehill",
+     "name": "Code Generation Agents",
+     "summary": "A comprehensive collection of specialized AI agents for code generation, development workflows, and project management. While originally designed for Claude Code, these agent specifications are…",
+     "gated": false,
+     "rows": null,
+     "bytes": 111355,
+     "license": "apache-2.0",
+     "modified": "2025-09-15"
+    },
+    {
+     "id": "danielrosehill/Geopolitical-System-Prompts",
+     "type": "System prompts",
+     "owner": "danielrosehill",
+     "name": "Geopolitical System Prompts",
+     "summary": "A collection of system prompts for AI assistants specialized in geopolitical analysis. These prompts enable AI systems to provide structured analysis, reporting, and insights across various aspects…",
+     "gated": false,
+     "rows": null,
+     "bytes": 17099,
+     "license": "apache-2.0",
+     "modified": "2025-04-09"
+    },
+    {
+     "id": "danielrosehill/Long-Prompt-Experiment",
+     "type": "Experiment",
+     "owner": "danielrosehill",
+     "name": "Long Prompt Experiment",
+     "summary": "I conducted this experiment to investigate the impact of prompt structure and optimization on LLM performance, specifically testing whether quality and organization matter more than raw prompt length…",
+     "gated": false,
+     "rows": 1,
+     "bytes": 12317400,
+     "license": "mit",
+     "modified": "2026-10-30"
+    },
+    {
+     "id": "danielrosehill/Shakespearean-Text-Transformation-Prompts",
+     "type": "Text prompts",
+     "owner": "danielrosehill",
+     "name": "Shakespearean Text Transformation Prompts",
+     "summary": "Welcome to what might be the internet's largest collection of prompts for rewriting text in Shakespearean English! This repository contains a variety of prompts designed to transform modern text into…",
+     "gated": false,
+     "rows": 1,
+     "bytes": 1054346,
+     "license": "apache-2.0",
+     "modified": "2025-04-21"
+    },
+    {
+     "id": "danielrosehill/Single-Prompt-Book",
+     "type": "Experiment",
+     "owner": "danielrosehill",
+     "name": "Single Prompt Book",
+     "summary": "An experiment in whether AI can write a book from just one prompt.",
+     "gated": false,
+     "rows": null,
+     "bytes": 962981,
+     "license": "apache-2.0",
+     "modified": "2025-04-09"
+    },
+    {
+     "id": "danielrosehill/Speech-To-Text-System-Prompts-2",
+     "type": "System prompts",
+     "owner": "danielrosehill",
+     "name": "Speech-To-Text System Prompts",
+     "summary": "This repository provides a collection of system prompts designed to transform and refine text captured using speech-to-text technologies.",
+     "gated": false,
+     "rows": 2,
+     "bytes": 384892,
+     "license": "apache-2.0",
+     "modified": "2025-04-09"
+    },
+    {
+     "id": "danielrosehill/System-Prompt-Library",
+     "type": "System prompts",
+     "owner": "danielrosehill",
+     "name": "System Prompt Library",
+     "summary": "This repository contains a comprehensive, up-to-date library of system prompts for AI systems and autonomous agents, started on May 27th, 2025.",
+     "gated": false,
+     "rows": 0,
+     "bytes": 3956580,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-03"
+    },
+    {
+     "id": "danielrosehill/Text-Transformation-Prompts-300525",
+     "type": "Text prompts",
+     "owner": "danielrosehill",
+     "name": "Text Transformation Prompts",
+     "summary": "A comprehensive collection of text transformation prompts for reformatting dictated text into various formats, styles, and structures.",
+     "gated": false,
+     "rows": 415,
+     "bytes": 307128,
+     "license": "cc-by-nd-4.0",
+     "modified": "2025-05-29"
+    },
+    {
+     "id": "danielrosehill/Writing-System-Prompts",
+     "type": "System prompts",
+     "owner": "danielrosehill",
+     "name": "Writing System Prompts",
+     "summary": "This is a collection of system prompts derived from my larger collection of system prompts. The commonality here is that these system prompts are intended for assistance related to writing…",
+     "gated": false,
+     "rows": 0,
+     "bytes": 173568,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-03"
+    }
+   ]
+  },
+  {
+   "slug": "mwp",
+   "tab": "My Weird Prompts",
+   "name": "My Weird Prompts",
+   "datasets": [
+    {
+     "id": "My-Weird-Prompts/Corn-The-Sloth",
+     "type": "Images",
+     "owner": "My-Weird-Prompts",
+     "name": "Corn The Sloth",
+     "summary": "This repository contains a small collection of images featuring an adorable plush sloth named Cornelius.",
+     "gated": false,
+     "rows": 109,
+     "bytes": 55198627,
+     "license": "artistic-2.0",
+     "modified": "2026-10-24"
+    },
+    {
+     "id": "My-Weird-Prompts/Corn-Training-Set",
+     "type": "Images",
+     "owner": "My-Weird-Prompts",
+     "name": "Corn Training Set",
+     "summary": "This repository is another image collection of images of a stuffed sloth that I am fine tuning for a custom image generation model for this particular character.",
+     "gated": false,
+     "rows": 118,
+     "bytes": 45110952,
+     "license": "apache-2.0",
+     "modified": "2026-10-24"
+    },
+    {
+     "id": "My-Weird-Prompts/episodes-eval",
+     "type": "Evaluation",
+     "owner": "My-Weird-Prompts",
+     "name": "Episodes Eval",
+     "summary": "212 episodes sampled from My-Weird-Prompts/episodes (revision 5fb2e59) for evaluating language models on long-form dialogue writing.",
+     "gated": false,
+     "rows": 212,
+     "bytes": 3190519,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-19"
+    },
+    {
+     "id": "My-Weird-Prompts/Eval-Prompts",
+     "type": "Evaluation",
+     "owner": "My-Weird-Prompts",
+     "name": "Eval Prompts",
+     "summary": "A fixed set of 10 real prompts from the My Weird Prompts podcast, used to compare script-writer models on identical input. Each prompt is verbatim as Daniel Rosehill sent it to the production…",
+     "gated": false,
+     "rows": 10,
+     "bytes": 22165,
+     "license": "cc-by-4.0",
+     "modified": "2026-11-02"
+    },
+    {
+     "id": "My-Weird-Prompts/Model-Chronicle",
+     "type": "Model history",
+     "owner": "My-Weird-Prompts",
+     "name": "Model Chronicle",
+     "summary": "My Weird Prompts is an AI-generated podcast: every episode's dialogue is written by a large language model, and the model that writes it has changed many times since November 2025. The back catalogue…",
+     "gated": false,
+     "rows": 59,
+     "bytes": 1442364,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-24"
+    },
+    {
+     "id": "My-Weird-Prompts/mwp-censorship-results",
+     "type": "Evaluation",
+     "owner": "My-Weird-Prompts",
+     "name": "MWP Censorship Results",
+     "summary": "This is the censorship track of MWP-Script. It is published separately because, for a show whose hosts are meant to speak frankly, a model that refuses or is filtered is disqualified whatever its…",
+     "gated": false,
+     "rows": 2654,
+     "bytes": 695656,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-19"
+    },
+    {
+     "id": "My-Weird-Prompts/MWP-Script-Results",
+     "type": "Evaluation",
+     "owner": "My-Weird-Prompts",
+     "name": "MWP Script Results",
+     "summary": "Results of MWP-Script, a benchmark for writing long-form, two-host podcast dialogue. It was built for the AI-written podcast My Weird Prompts.",
+     "gated": false,
+     "rows": 102,
+     "bytes": 3543296,
+     "license": "cc-by-4.0",
+     "modified": "2026-10-19"
+    }
+   ]
+  }
+ ],
+ "inside": [
+  {
+   "id": "Hebrish/Acronym-Sentences",
+   "name": "Acronym Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/AI-Interactions-Sentence-Bank",
+   "name": "AI Interactions Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/AI-Interactions-Sentences",
+   "name": "AI Interactions Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/AI-Tech-Vocabulary-Sentence-Bank",
+   "name": "AI Tech Vocabulary Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/American-Centrism-Eval",
+   "name": "American Centrism Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/ASR-WPM-And-Background-Noise-Eval",
+   "name": "ASR WPM And Background Noise Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Audio-Understanding-Bitrate-Eval-0426",
+   "name": "Audio Understanding Bitrate Eval 0426",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Audio-Understanding-Test-Set",
+   "name": "Audio Understanding Test Set",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Brand-Name-Sentences",
+   "name": "Brand Name Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Conspiracy-Endorsement-Eval",
+   "name": "Conspiracy Endorsement Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Conspiracy-Endorsement-Eval-Runs",
+   "name": "Conspiracy Endorsement Eval Runs",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Currency-Sentences",
+   "name": "Currency Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Disfluent-Read-Sentences",
+   "name": "Disfluent Read Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/English-Hebrew-Codeswitching-ASR-Benchmark",
+   "name": "English Hebrew Codeswitching ASR Benchmark",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/English-Hebrew-Codeswitching-Sentences",
+   "name": "English Hebrew Codeswitching Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/English-Hebrew-Mixed-Sentences",
+   "name": "English Hebrew Mixed Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "My-Weird-Prompts/episodes",
+   "name": "Episodes",
+   "nav": "mwp"
+  },
+  {
+   "id": "Hebrish/Example-Sentence-Bank",
+   "name": "Example Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "Hebrish/Expanded-Acronym-Sentences",
+   "name": "Expanded Acronym Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/General-English-Sentences",
+   "name": "General English Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Hard-ASR-Sentences",
+   "name": "Hard ASR Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Hebrew-Image-Eval-111225",
+   "name": "Hebrew Image Eval 111225",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Hebrish-TTS-Voice-Recordings",
+   "name": "Hebrish TTS Voice Recordings",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Hebrish-Validation",
+   "name": "Hebrish Validation",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Homophone-Context-Sentence-Bank",
+   "name": "Homophone Context Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/Homophone-Context-Sentences",
+   "name": "Homophone Context Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Irish-English-Sentences",
+   "name": "Irish English Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Israel-Palestine-Framing-Eval",
+   "name": "Israel Palestine Framing Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Israel-Palestine-Framing-Eval-Runs",
+   "name": "Israel Palestine Framing Eval Runs",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Israeli-Name-Sentences",
+   "name": "Israeli Name Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Jewish-Religious-Terminology-Sentences",
+   "name": "Jewish Religious Terminology Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Judaica-Sentences",
+   "name": "Judaica Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Linux-Admin-Sentence-Bank",
+   "name": "Linux Admin Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/Linux-Admin-Sentences",
+   "name": "Linux Admin Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Noisy-Condition-Sentences",
+   "name": "Noisy Condition Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Noisy-Voice-Notes",
+   "name": "Noisy Voice Notes",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/October-7-Memorial-Sentence-Bank",
+   "name": "October 7 Memorial Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/Paragraph-Dictation-Passages",
+   "name": "Paragraph Dictation Passages",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Parakeet-Dictation-Cleanup-Eval",
+   "name": "Parakeet Dictation Cleanup Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Personal-Vocabulary-Sentence-Bank",
+   "name": "Personal Vocabulary Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/Personal-Vocabulary-Sentences",
+   "name": "Personal Vocabulary Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Podcast-ASR-Evaluation",
+   "name": "Podcast ASR Evaluation",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Pronunciation-Variants",
+   "name": "Pronunciation Variants",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Proofed-ASR-Cleanup-Training-Data",
+   "name": "Proofed ASR Cleanup Training Data",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Propaganda-Bait-Eval",
+   "name": "Propaganda Bait Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Propaganda-Bait-Eval-Runs",
+   "name": "Propaganda Bait Eval Runs",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Read-Aloud-Voice-Sentences",
+   "name": "Read Aloud Voice Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Repeated-Word-Sentences",
+   "name": "Repeated Word Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Self-Correction-Sentences",
+   "name": "Self Correction Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Singing-Sentences",
+   "name": "Singing Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Small-STT-Eval-Audio-Dataset",
+   "name": "Small STT Eval Audio Dataset",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Smart-Home-Sentence-Bank",
+   "name": "Smart Home Sentence Bank",
+   "nav": "sentence"
+  },
+  {
+   "id": "danielrosehill/Smart-Home-Sentences",
+   "name": "Smart Home Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Spoken-Aside-Sentences",
+   "name": "Spoken Aside Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/STT-Fine-Tune-Eval-101225",
+   "name": "STT Fine Tune Eval 101225",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/STT-Voice-Notes-Evals",
+   "name": "STT Voice Notes Evals",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Tech-Graphics-Benchmark",
+   "name": "Tech Graphics Benchmark",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Tech-Graphics-Benchmark-Ratings",
+   "name": "Tech Graphics Benchmark Ratings",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Tech-Graphics-Benchmark-Runs",
+   "name": "Tech Graphics Benchmark Runs",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Tech-Sentences",
+   "name": "Tech Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Tech-Sentences-For-ASR-Training",
+   "name": "Tech Sentences For ASR Training",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Text-To-Image-Test-Prompts",
+   "name": "Text To Image Test Prompts",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/The-PC-Bench",
+   "name": "The PC Bench",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "My-Weird-Prompts/transcripts",
+   "name": "Transcripts",
+   "nav": "mwp"
+  },
+  {
+   "id": "danielrosehill/US-Partisan-Framing-Eval",
+   "name": "US Partisan Framing Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "danielrosehill/Voice-Command-Sentences",
+   "name": "Voice Command Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Weird-Voice-Sentences",
+   "name": "Weird Voice Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "danielrosehill/Whisper-Fine-Tune-One-Shot-Eval",
+   "name": "Whisper Fine Tune One Shot Eval",
+   "nav": "benchmarks"
+  },
+  {
+   "id": "Hebrish/Whispered-Sentences",
+   "name": "Whispered Sentences",
+   "nav": "audio"
+  },
+  {
+   "id": "Hebrish/Yiddish-Sentences",
+   "name": "Yiddish Sentences",
+   "nav": "audio"
+  }
+ ]
+}

index.html

@@ -1,19 +1,18 @@
 <!doctype html>
-<html>
-	<head>
-		<meta charset="utf-8" />
-		<meta name="viewport" content="width=device-width" />
-		<title>My static Space</title>
-		<link rel="stylesheet" href="style.css" />
-	</head>
-	<body>
-		<div class="card">
-			<h1>Welcome to your static Space!</h1>
-			<p>You can modify this app directly by editing <i>index.html</i> in the Files and versions tab.</p>
-			<p>
-				Also don't forget to check the
-				<a href="https://huggingface.co/docs/hub/spaces" target="_blank">Spaces documentation</a>.
-			</p>
-		</div>
-	</body>
-</html>
+<html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1">
+<title>Dataset Navigator</title><link rel="stylesheet" href="style.css">
+<style>nav.tabs{flex-wrap:wrap}nav.tabs a{font-size:1.02rem}.facts{margin:auto 0 0;padding-top:8px;border-top:1px solid var(--line);display:grid;grid-template-columns:auto 1fr;gap:3px 12px;font-size:.84rem}.facts dt{color:var(--muted)}.facts dd{margin:0;overflow-wrap:anywhere}.count{font-size:1.5rem;font-weight:600;line-height:1.1;margin-top:auto}.count span{font-size:.9rem;font-weight:400;color:var(--muted)}</style></head>
+<body><main>
+<h1>Dataset Navigator</h1>
+<p class="lede">Daniel Rosehill's public datasets in one place, from the danielrosehill account and the Hebrish and My Weird Prompts organisations. Start with a specialist navigator, or use the topic tabs for everything else. Search reaches all of them.</p>
+<nav class="links">
+<a href="https://huggingface.co/spaces/danielrosehill/Start-Here" target="_blank" rel="noopener">Start Here</a>
+<a href="https://huggingface.co/spaces/danielrosehill/Model-Navigator" target="_blank" rel="noopener">Model Navigator</a>
+<a href="https://huggingface.co/datasets/Hebrish/Dataset-Index" target="_blank" rel="noopener">Hebrish dataset index</a>
+<a href="https://ai.danielrosehill.com/collections" target="_blank" rel="noopener">Collections</a>
+</nav>
+<div class="controls"><input id="q" type="search" placeholder="Search all datasets" aria-label="Search all datasets"></div>
+<nav class="tabs" id="tabs" aria-label="Sections"></nav>
+<div id="list"></div>
+<footer><span id="count"></span> Data from the Hugging Face Hub, built <span id="built"></span>. Row counts come from the dataset viewer and are left out where it can't count a repo's files.</footer>
+</main><script src="app.js"></script></body></html>

style.css

@@ -1,28 +1,31 @@
-body {
-	padding: 2rem;
-	font-family: -apple-system, BlinkMacSystemFont, "Arial", sans-serif;
-}
-
-h1 {
-	font-size: 16px;
-	margin-top: 0;
-}
-
-p {
-	color: rgb(107, 114, 128);
-	font-size: 15px;
-	margin-bottom: 10px;
-	margin-top: 5px;
-}
-
-.card {
-	max-width: 620px;
-	margin: 0 auto;
-	padding: 16px;
-	border: 1px solid lightgray;
-	border-radius: 16px;
-}
-
-.card p:last-child {
-	margin-bottom: 0;
-}
+:root{--bg:#f7f7f4;--card:#fff;--ink:#1d2327;--muted:#5d676e;--line:#dfe2dd;--accent:#0b6e69;--soft:#e3f1ef;--gate:#8a5a00;--gatebg:#fbf0d9}
+*{box-sizing:border-box}
+body{margin:0;background:var(--bg);color:var(--ink);font:16px/1.5 system-ui,-apple-system,"Segoe UI",sans-serif}
+main{max-width:1080px;margin:0 auto;padding:32px 20px 56px}
+a{color:var(--accent)}
+h1{font-size:2rem;line-height:1.2;margin:0 0 8px}
+h2{font-size:1.15rem;margin:32px 0 12px}
+.lede{color:var(--muted);max-width:72ch;margin:0 0 14px}
+nav.links{display:flex;flex-wrap:wrap;gap:8px 18px;margin:0 0 20px;font-size:.95rem}
+nav.tabs{display:flex;gap:4px;border-bottom:1px solid var(--line);margin:0 0 22px}
+nav.tabs a{padding:8px 14px;text-decoration:none;color:var(--muted);border-bottom:3px solid transparent}
+nav.tabs a[aria-current]{color:var(--ink);border-color:var(--accent);font-weight:600}
+.stats{display:flex;flex-wrap:wrap;gap:10px 28px;margin:0 0 20px;color:var(--muted);font-size:.95rem}
+.stats b{color:var(--ink);font-size:1.15rem}
+.controls{display:flex;flex-wrap:wrap;gap:10px;align-items:center;margin-bottom:6px}
+input[type=search]{flex:1 1 240px;padding:9px 12px;border:1px solid var(--line);border-radius:6px;font:inherit;background:#fff}
+select{padding:9px 10px;border:1px solid var(--line);border-radius:6px;font:inherit;background:#fff}
+.grid{display:grid;grid-template-columns:repeat(auto-fill,minmax(310px,1fr));gap:14px}
+.card{background:var(--card);border:1px solid var(--line);border-radius:8px;padding:14px 16px;display:flex;flex-direction:column;gap:8px}
+.card.master{grid-column:1/-1;border-color:var(--accent);border-width:2px}
+.who{display:flex;align-items:center;gap:8px;font-size:.82rem;color:var(--muted)}
+.who img{width:24px;height:24px;border-radius:5px;object-fit:cover}
+.card h3{margin:0;font-size:1.05rem;line-height:1.3}
+.card h3 a{text-decoration:none;color:var(--ink)}
+.card h3 a:hover{color:var(--accent);text-decoration:underline}
+.card p{margin:0;color:var(--muted);font-size:.93rem}
+.meta{display:flex;flex-wrap:wrap;gap:6px 12px;font-size:.84rem;color:var(--muted);margin-top:auto;align-items:center}
+.badge{background:var(--gatebg);color:var(--gate);padding:1px 7px;border-radius:4px}
+.empty{color:var(--muted);padding:24px 0}
+footer{margin-top:40px;padding-top:14px;border-top:1px solid var(--line);color:var(--muted);font-size:.86rem}
+@media(max-width:520px){h1{font-size:1.6rem}main{padding:22px 14px 40px}}