INDEX

Explanations

application

np_max-act · gemini-2.0-flash

The neuron detects legal‐procedural action terms (e.g. applying, filing, appealing, requesting review).

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

謝

-0.07

 appears

-0.06

显

-0.06

Too

-0.06

 homosexual

-0.06

 ring

-0.06

 operation

-0.06

problem

-0.06

 spreading

-0.06

 stunt

-0.06

POSITIVE LOGITS

 flexDirection

0.07

 aprend

0.07

 федера

0.07

 відбу

0.07

 strtok

0.06

 Sexe

0.06

 đèn

0.06

 FormGroup

0.06

 металли

0.06

Activations Density 0.012%

application

The neuron detects legal‐procedural action terms (e.g. applying, filing, appealing, requesting review).

No Comments

No Known Activations

application

The neuron detects legal‐procedural action terms (e.g. applying, filing, appealing, requesting review).

No Comments

No Known Activations