INDEX

Explanations

cause and effect

np_max-act · gemini-2.0-flash

The neuron selectively responds to small function words—especially prepositions like “for,” “of,” and “in”—that introduce purposes or relationships in technical headings and descriptions.

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

telefono

-0.07

 삭제

-0.07

ノ

-0.07

 Holding

-0.07

 th�

-0.06

 midway

-0.06

 violently

-0.06

Clo

-0.06

silver

-0.06

-split

-0.06

POSITIVE LOGITS

�ng

0.07

IRM

0.06

_DDR

0.06

CONTENT

0.06

lip

0.06

 armor

0.06

;margin

0.06

number

0.06

(SYS

0.06

Hmm

0.06

Activations Density 0.051%

cause and effect

The neuron selectively responds to small function words—especially prepositions like “for,” “of,” and “in”—that introduce purposes or relationships in technical headings and descriptions.

No Comments

No Known Activations