INDEX

Explanations

period

np_max-act · gemini-2.0-flash

This neuron lights up on the court‐opinion header information—docket numbers, party names, “v.,” and similar case‐caption metadata.

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

Wak

-0.07

.StatusBadRequest

-0.07

 impress

-0.07

 اع

-0.06

 olmaz

-0.06

Cr

-0.06

[^

-0.06

leave

-0.06

Maz

-0.06

出し

-0.06

POSITIVE LOGITS

uales

0.06

ivní

0.06

atorial

0.06

BIG

0.06

]]↵↵

0.06

Narr

0.06

 DISABLE

0.06

 กล

0.06

 dataIndex

0.06

scaled

0.06

Activations Density 0.012%

period

This neuron lights up on the court‐opinion header information—docket numbers, party names, “v.,” and similar case‐caption metadata.

No Comments

No Known Activations