INDEX

Explanations

keywords associated with transactions and products

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

noun phrase completion

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

interopRequire

-0.57

SharedCtor

-0.55

GOTREF

-0.55

setVerticalGroup

-0.52

OCCURRED

-0.51

BufferException

-0.46

JspWriter

-0.46

adaptiveStyles

-0.45

archiviato

-0.45

LookAnd

-0.44

POSITIVE LOGITS

TagMode

0.55

 esternos

0.49

uttgart

0.47

ACHUSET

0.46

straints

0.45

Життєпис

0.41

TokenNameEQUAL

0.40

delige

0.40

trice

0.40

 fixing

0.40

Activations Density 0.006%

keywords associated with transactions and products

noun phrase completion

No Comments

No Known Activations