INDEX

Explanations

on

np_max-act · gemini-2.0-flash

tokens that appear in headings, titles, links or other prominent document-level metadata (e.g., subject lines, URLs, proper‑names).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

New Auto-Interp

Configuration

andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ជ

-0.08

铬

-0.07

wil

-0.07

锆

-0.07

 consolidation

-0.07

 Million

-0.07

 exploiting

-0.07

 pollen

-0.07

蓬勃发展

-0.07

KB

-0.07

POSITIVE LOGITS

 phạm

0.07

 너희

0.07

 작성

0.07

sø

0.07

 Petsc

0.07

(*)

0.07

 שאני

0.07

(ray

0.07

);}↵

0.07

 KeyError

0.07

Activations Density 0.281%

on

tokens that appear in headings, titles, links or other prominent document-level metadata (e.g., subject lines, URLs, proper‑names).

No Comments

No Known Activations