INDEX

Explanations

possibility

np_max-act · gemini-2.0-flash

The neuron consistently activates on the word “possibility” (as in “possibility of such damage”) in software‐license disclaimer blocks.

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

VISIBLE

-0.07

(w

-0.07

哦

-0.06

 mandate

-0.06

($(

-0.06

.ColumnHeadersHeightSizeMode

-0.06

 économ

-0.06

імеч

-0.06

washing

-0.06

 PendingIntent

-0.06

POSITIVE LOGITS

_self

0.07

 POSSIBILITY

0.07

่ได

0.06

追

0.06

 tram

0.06

 Residence

0.06

_DOC

0.06

 اضافه

0.06

 SCIP

0.06

VIP

0.06

Activations Density 0.000%

possibility

The neuron consistently activates on the word “possibility” (as in “possibility of such damage”) in software‐license disclaimer blocks.

No Comments

No Known Activations