INDEX

Explanations

accusations

np_max-act · gemini-2.0-flash

The neuron specializes in spotting words and phrases tied to legal proceedings—like trials, charges, convictions, counts, and courtroom actions.

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

lectual

-0.06

 Decision

-0.06

Me

-0.06

ipple

-0.06

 milan

-0.06

	draw

-0.06

.getDouble

-0.06

_uc

-0.06

 hierarchical

-0.06

isure

-0.06

POSITIVE LOGITS

azio

0.07

 petites

0.07

 pylab

0.07

(vm

0.07

_HANDLER

0.06

 palp

0.06

(RuntimeObject

0.06

.rest

0.06

ři

0.06

:create

0.06

Activations Density 0.028%

accusations

The neuron specializes in spotting words and phrases tied to legal proceedings—like trials, charges, convictions, counts, and courtroom actions.

No Comments

No Known Activations