INDEX

Explanations

instances of the beginning of new document sections or paragraphs

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.64

</em>

-0.56

↵

-0.46

สาม

-0.46

-0.45

-0.44

fo

-0.44

stat

-0.43

it

-0.43

SL

-0.43

POSITIVE LOGITS

تقاوى

1.25

rungsseite

1.02

InputBorder

1.00

GEBURTSDATUM

0.99

AnchorStyles

0.98

 transfieras

0.98

RectangleBorder

0.97

 BIBSYS

0.94

 Normdatei

0.92

RUnlock

0.92

Activations Density 0.075%

instances of the beginning of new document sections or paragraphs

No Comments

No Known Activations