INDEX

Explanations

phrases indicating exceptions or limitations

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

"...but are not limited to..."

np_acts-logits-general · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

CrossRef

-0.47

dian

-0.41

ươ

-0.41

 Dietz

-0.39

idist

-0.39

Bud

-0.37

roman

-0.36

越

-0.36

LookAnd

-0.36

 onward

-0.36

POSITIVE LOGITS

 limited

1.59

 confined

1.52

limited

1.40

 confine

1.35

Limited

1.34

 restricted

1.32

confined

1.30

 limitado

1.30

 terbatas

1.27

 limitada

1.23

Activations Density 0.250%

phrases indicating exceptions or limitations

"...but are not limited to..."

No Comments

No Known Activations