PulseAugur
EN
LIVE 15:38:56

New 13B language model "Talkie" trained on pre-1931 English text

By PulseAugur Editorial · [1 sources] ·

A new language model named "Talkie" has been released, featuring 13 billion parameters and trained on 260 billion tokens of English text predating 1931. An instruction-tuned version of Talkie incorporates historical etiquette, letters, encyclopedias, and poetry to provide a controlled perspective on temporal training data. AI

IMPACT This model's unique training data could offer new insights into historical language patterns and their impact on AI.

RANK_REASON The cluster describes the release of a new language model with specific training data characteristics, fitting the research category. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Mastodon — mastodon.social →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

New 13B language model "Talkie" trained on pre-1931 English text

How we ranked this

Signal score
11 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster describes the release of a new language model with specific training data characteristics, fitting the research category. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Mastodon — mastodon.social TIER_1 English(EN) · [email protected] ·

    "Talkie" is a 13B language model trained on 260B tokens of pre-1931 English text. Its instruction-tuned variant uses historical etiquette, letters, encyclopedia

    "Talkie" is a 13B language model trained on 260B tokens of pre-1931 English text. Its instruction-tuned variant uses historical etiquette, letters, encyclopedias and poetry, offering a controlled view of temporal training data. # LLM # AI # Talkie https:// talkie-lm.com/introduci…

    LINKS talkie-lm.com/introducing-talkie

RELATED ENTITIES

RELATED TOPICS