← PTS Pivotal Token SearchCONTENT HISTORY
WHAT CHANGED
added
Update to PTS Pivotal Token Search
Snapshot Oct 27, 2026 · 00:49 UTC · version 0.1.0
Package or technical metadata updated
Package or technical metadata updated
Observed in package metadata. These changes alone do not establish a new customer-facing feature.
Save PTS Pivotal Token Search and check future changes in Following.
Up to 10 plugins free · no card required.
All fields and original values.
Is v2 template
Before
Oct 24, 2026 · 13:36 UTC
Not present
After
Oct 27, 2026 · 00:49 UTC
false
Compare saved observations
Download comparison JSONFull technical diff · 1 changed fields
added /is_v2_template
BEFORE
Field was absent
AFTER
false
Full snapshot data
{
"canonical_app_id": null,
"connector_id": null,
"created_at": "2026-10-01T05:32:30.722251Z",
"discoverability": "LISTED",
"id": "plugins_6abdee7dcfb88191a5353b4cb83df8cc",
"is_template": false,
"is_template_instance": false,
"is_v2_template": false,
"name": "pts-pivotal-token-search",
"release": {
"app_ids": [],
"app_manifest": null,
"app_templates": [],
"description": "Guides you through finding the reasoning events that change a language model's outcome with the open-source PTS framework.",
"display_name": "PTS Pivotal Token Search",
"id": "pluginrel_0a4a704ad210819181c6fdf1c0656213",
"interface": {
"brand_color": "#7C3AED",
"capabilities": [],
"category": "Scientific Research",
"composer_icon_dark_url": null,
"composer_icon_url": "https://files.openai.com/content?id=file_0000000069c48207ae40800f5adedb2c",
"default_prompt": "Explain what PTS finds and when to use it",
"default_prompts": [
"Explain what PTS finds and when to use it",
"Plan a small PTS search on my task",
"Help me read my PTS events file"
],
"developer_name": "PTS (codelion)",
"logo_url": "https://files.openai.com/content?id=file_000000000eb082098a235068264d7457",
"logo_url_dark": null,
"long_description": "Guides you through PTS (Pivotal Token Search), an open-source framework for finding the reasoning events that change whether a language model solves a task. The workflow explains the three scales PTS searches (latent workspace meta-tokens, emitted pivotal tokens, and reasoning sentences called thought anchors), installs PTS from its README after asking you, and shows how to run a search on a task with a small local model. It explains how to read the output: probability shifts, the shared category labels, and why latent readout scores are not comparable to probability changes. It covers the hosted visualizer, linking events into chains, and exporting results as datasets for preference tuning or steering. It states the limits plainly: searches are expensive because they sample many continuations, results are observational unless you run an intervention with controls, and findings depend on the task, the model and the sampling settings. This plugin contains instructions only; everything runs on your own machine.",
"plugin_category_id": "scientific research",
"privacy_policy_url": "https://meragpt.com/privacy",
"screenshot_urls": [],
"short_description": "Find pivotal reasoning events",
"terms_of_service_url": "https://meragpt.com/terms",
"website_url": "https://github.com/codelion/pts"
},
"keywords": [
"interpretability",
"reasoning",
"pivotal tokens",
"thought anchors",
"datasets"
],
"mcp_servers": [],
"onboarding_skill_name": null,
"requires_local_executor": false,
"skills": [
{
"description": "Use PTS (Pivotal Token Search) to find the tokens, sentences and latent concepts that change whether a language model solves a task, read the resulting events file, explore it in the visualizer, and export it as datasets. Use when the user wants to analyze reasoning steps of a small local model, interpret a PTS events file, or plan a search.",
"interface": {
"brand_color": null,
"default_prompt": null,
"display_name": "pts-pivotal-token-search",
"icon_large_url": null,
"icon_small_url": null,
"iconography": "radar",
"short_description": "Use PTS (Pivotal Token Search) to find the tokens, sentences and latent concepts that change whether a language model solves a task, read the resulting events file, explore it in the visualizer, and export it as datasets. Use when the user wants to analyze reasoning steps of a small local model, interpret a PTS events file, or plan a search."
},
"name": "pts-pivotal-token-search",
"plugin_release_skill_id": "pluginrsk_6abdef3833e4819194725bdec41e0c49"
}
],
"version": "0.1.0"
},
"scope": "GLOBAL",
"share_url": "https://chatgpt.com/plugins/plugins_6abdee7dcfb88191a5353b4cb83df8cc?open_in_app",
"status": "ENABLED"
}SHA-256 of public snapshot: cfd637a47a980cd249c14c103f555a702fd7091641d8faa1863c57fc570a11e1