INDEX

Explanations

specific years or numerical data related to scientific or technical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

zugehen

-0.38

 Saltar

-0.35

 Grüsse

-0.35

 Administrativna

-0.33

 złot

-0.32

 tightly

-0.31

 directly

-0.31

 Dichtung

-0.31

 hatta

-0.31

 Kontroll

-0.31

POSITIVE LOGITS

guz

1.00

 ProtoMessage

0.63

 Roskov

0.62

 Guin

0.57

__*/

0.50

 combineReducers

0.49

forRoot

0.48

PerPage

0.48

WSGI

0.47

nextSibling

0.46

Activations Density 0.000%

specific years or numerical data related to scientific or technical contexts

No Comments

No Known Activations