INDEX

Explanations

references to names, events, or items related to food, specifically hot dogs and other sandwiches, as well as academic or scientific terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

Personendaten

-0.59

httphttps

-0.56

 disambiguazione

-0.54

 beginnetje

-0.50

 שוליים

-0.48

 autorytatywna

-0.48

 المعيارى

-0.48

UnsafeEnabled

-0.47

MENAFN

-0.47

AnchorTagHelper

-0.46

POSITIVE LOGITS

ck

0.99

CK

0.94

CK

0.94

ck

0.88

ek

0.84

Ck

0.84

pk

0.84

PK

0.83

DK

0.82

ak

0.80

Activations Density 1.533%

references to names, events, or items related to food, specifically hot dogs and other sandwiches, as well as academic or scientific terms

No Comments

No Known Activations