INDEX

Explanations

verbal

np_max-act · gemini-2.0-flash

The neuron is keyed to spotting the word “verbal” (as in “verbal communication”).

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

makt

-0.07

"Now

-0.07

 pesos

-0.07

工程

-0.06

該

-0.06

 FormData

-0.06

 Mais

-0.06

наслідок

-0.06

 spoiled

-0.06

排名

-0.06

POSITIVE LOGITS

 verbal

0.15

 verbally

0.12

 utter

0.08

ver

0.07

언

0.07

verbs

0.07

Verb

0.07

Url

0.07

 Verb

0.07

HTTPRequest

0.07

Activations Density 0.005%

verbal

The neuron is keyed to spotting the word “verbal” (as in “verbal communication”).

No Comments

No Known Activations