INDEX

Explanations

e.g

np_max-act · gemini-2.0-flash

The neuron is triggered by the abbreviation “e.g.” (i.e. the marker used to introduce examples).

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

 реж

-0.07

มหานคร

-0.07

 JSBracketAccess

-0.07

_STEP

-0.07

ائ

-0.06

 palabra

-0.06

ﾘ

-0.06

 powerhouse

-0.06

.getUserName

-0.06

 CUDA

-0.06

POSITIVE LOGITS

Exited

0.07

 deducted

0.06

ーブル

0.06

ustainable

0.06

 bitter

0.06

SH

0.06

….

0.06

Service

0.06

 españ

0.06

Activations Density 0.034%

e.g

The neuron is triggered by the abbreviation “e.g.” (i.e. the marker used to introduce examples).

No Comments

No Known Activations