INDEX

Explanations

Legal affirmations/grants

np_max-act · gemini-2.0-flash

This neuron detects judicial disposition terms in legal opinions (e.g., “AFFIRM,” “REVERSE,” “DENIED,” “DISMISSED,” “OPINION”).

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

 heaters

-0.07

 Tank

-0.07

 conceive

-0.07

Arthur

-0.06

 chloride

-0.06

approval

-0.06

 pens

-0.06

NY

-0.06

 Enabled

-0.06

も

-0.06

POSITIVE LOGITS

种

0.07

>').

0.06

.partial

0.06

 Поп

0.06

[row

0.06

 priorit

0.06

 Komm

0.06

 brunette

0.06

 mediaPlayer

0.06

.AppendLine

0.06

Activations Density 0.003%

Legal affirmations/grants

This neuron detects judicial disposition terms in legal opinions (e.g., “AFFIRM,” “REVERSE,” “DENIED,” “DISMISSED,” “OPINION”).

No Comments

No Known Activations