INDEX

Explanations

expressions of strong, aggressive language or expletives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

تقاوى

-0.53

 lenker

-0.52

)•

-0.52

Personendaten

-0.47

AndEndTag

-0.46

Diweddarwch

-0.45

 ſever

-0.42

 فريبيس

-0.41

MemoryWarning

-0.40

 itſelf

-0.40

POSITIVE LOGITS

 fucking

2.98

 goddamn

2.64

 fuckin

2.64

 fuck

2.63

fucking

2.53

 Fucking

2.52

 FUCKING

2.44

 damn

2.39

 Fuck

2.38

Fucking

2.38

Activations Density 0.805%

expressions of strong, aggressive language or expletives

No Comments

No Known Activations