INDEX

Explanations

period

np_max-act · gemini-2.0-flash

The neuron spots section‐heading and other high‐level structural tokens in a biographical article (e.g. “Biography,” “Art,” “In,” “She,” “He,” “and” at the start of new narrative sections).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

("\"

-0.06

فال

-0.06

conversion

-0.06

seven

-0.06

opt

-0.06

ajar

-0.06

てい

-0.06

 registrar

-0.06

لب

-0.06

piece

-0.06

POSITIVE LOGITS

 TODO

0.08

(My

0.07

 undermined

0.06

Product

0.06

PEnd

0.06

_CENTER

0.06

еры

0.06

 Consum

0.06

!");↵

0.06

 cuales

0.06

Activations Density 0.037%

period

The neuron spots section‐heading and other high‐level structural tokens in a biographical article (e.g. “Biography,” “Art,” “In,” “She,” “He,” “and” at the start of new narrative sections).

No Comments

No Known Activations