INDEX

Explanations

want

np_max-act · gemini-2.0-flash

The neuron detects reader-directed call-to-action phrases, especially “if you want to …” 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

 principalColumn

-0.06

Stra

-0.06

 Matth

-0.06

форми

-0.06

伝

-0.06

 backgroundImage

-0.06

 لكن

-0.06

 Cort

-0.06

 tốc

-0.06

Terms

-0.06

POSITIVE LOGITS

 templ

0.07

 sandwich

0.07

insert

0.07

 Acting

0.07

_observer

0.06

 xhttp

0.06

_database

0.06

TO

0.06

 smashing

0.06

Để

0.06

Activations Density 0.020%

want

The neuron detects reader-directed call-to-action phrases, especially “if you want to …” constructions.

No Comments

No Known Activations