INDEX

Explanations

interpersonal interactions/speaking

np_max-act · gemini-2.0-flash

The neuron fires on occurrences of “use” (and its contexts like “they use” or “someone … use”), i.e. it detects the verb “use” in conditional/subject-verb constructions.

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

Theory

-0.06

σφ

-0.06

_activities

-0.06

 hamburger

-0.05

evším

-0.05

 Surg

-0.05

[/

-0.05

.uk

-0.05

्च

-0.05

 parody

-0.05

POSITIVE LOGITS

tır

0.07

 Họ

0.07

Bubble

0.06

 drains

0.06

adients

0.06

 عليه

0.06

 десяти

0.06

ूद

0.06

uspend

0.06

 область

0.06

Activations Density 0.007%

interpersonal interactions/speaking

The neuron fires on occurrences of “use” (and its contexts like “they use” or “someone … use”), i.e. it detects the verb “use” in conditional/subject-verb constructions.

No Comments

No Known Activations

interpersonal interactions/speaking

The neuron fires on occurrences of “use” (and its contexts like “they use” or “someone … use”), i.e. it detects the verb “use” in conditional/subject-verb constructions.

No Comments

No Known Activations