INDEX

Explanations

phrases indicating the absence of evidence or suggestions

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

implying absence or lack

np_acts-logits-general · gemini-2.0-flash

nothing or not anything

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

 ujednoznacz

-0.60

 AssemblyCulture

-0.58

Autoritní

-0.54

StructEnd

-0.49

SequentialGroup

-0.48

errHandler

-0.47

Personendaten

-0.47

FXMLLoader

-0.46

PageRoute

-0.46

expandindo

-0.46

POSITIVE LOGITS

 whatsoever

0.96

 none

0.84

 nothing

0.79

none

0.78

 None

0.76

 nimic

0.76

 żad

0.75

aucun

0.75

nothing

0.75

Nothing

0.75

Activations Density 0.625%

phrases indicating the absence of evidence or suggestions

implying absence or lack

nothing or not anything

No Comments

No Known Activations