INDEX

Explanations

instances of the word "to"

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

movement to a destination

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Antara

-0.55

modelBuilder

-0.55

 endblock

-0.53

adaptiveStyles

-0.52

المناصب

-0.50

awtextra

-0.50

KeySpec

-0.50

mimetype

-0.49

sizeCache

-0.49

chartInstance

-0.49

POSITIVE LOGITS

 goTo

0.46

goTo

0.44

atteindre

0.43

󠁣

0.41

encodeWith

0.40

bootstrapcdn

0.40

ICIAL

0.39

 Went

0.38

 Expédié

0.38

 desplazamiento

0.37

Activations Density 0.015%

instances of the word "to"

movement to a destination

No Comments

No Known Activations