INDEX

Explanations

phrases indicating causation or conditions of existence

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

to + verb

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

 <<<<<<<<<<<<<<

-0.69

̈́

-0.69

كويكب

-0.66

 שוליים

-0.65

RotationOrder

-0.65

+#+#

-0.60

SharedCtor

-0.60

Datuak

-0.60

ंदीखरीदारी

-0.59

findpost

-0.58

POSITIVE LOGITS

 esser

0.32

 étant

0.32

 being

0.31

 JsonResponse

0.31

 estar

0.30

事儿

0.30

 PATENT

0.29

 patent

0.29

 ότι

0.29

 escort

0.29

Activations Density 0.087%

phrases indicating causation or conditions of existence

to + verb

No Comments

No Known Activations