INDEX

Explanations

the

np_max-act · gemini-2.0-flash

The neuron strongly activates on words and phrases relating to sentencing—especially explicit mentions of “sentence,” “sentencing,” “Guidelines,” and analogous legal‐procedure terminology.

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

 save

-0.08

_Ad

-0.07

失败

-0.06

Blo

-0.06

'&

-0.06

 několik

-0.06

 sudoku

-0.06

重

-0.06

_ass

-0.06

Pep

-0.06

POSITIVE LOGITS

 Petersburg

0.07

蔵

0.07

icopter

0.06

 октября

0.06

().__

0.06

istrates

0.06

яви

0.06

istar

0.06

caster

0.06

istic

0.06

Activations Density 0.010%

the

The neuron strongly activates on words and phrases relating to sentencing—especially explicit mentions of “sentence,” “sentencing,” “Guidelines,” and analogous legal‐procedure terminology.

No Comments

No Known Activations