INDEX

Explanations

phrases indicating disrespect or negative consequences in professional contexts

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

Comes before punctuation

np_acts-logits-general · gemini-2.0-flash

referencing further details

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 JpaRepository

-0.34

Personendaten

-0.33

sträger

-0.32

origin

-0.32

 Enabling

-0.32

IUrlHelper

-0.31

 enables

-0.31

strut

-0.30

 ujednoznacz

-0.29

cof

-0.29

POSITIVE LOGITS

 للاسماء

0.62

 falsche

0.55

PyExc

0.54

UpInside

0.52

eventName

0.50

SharedCtor

0.50

AMAHA

0.50

 NSCoder

0.50

BufferException

0.49

yelitis

0.49

Activations Density 0.621%

phrases indicating disrespect or negative consequences in professional contexts

Comes before punctuation

referencing further details

No Comments

No Known Activations