INDEX

Explanations

phrases associated with intent, particularly those that indicate a desire or action to do something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

透

-0.42

create

-0.42

hability

-0.42

est

-0.40

ny

-0.40

have

-0.40

http

-0.40

esen

-0.39

的多

-0.39

-0.38

POSITIVE LOGITS

 nahilalakip

0.97

 nakalista

0.88

 surla

0.84

anskje

0.83

 behold

0.81

AndEndTag

0.80

 myſelf

0.79

 itſelf

0.76

 Anſ

0.76

 consider

0.76

Activations Density 0.164%

phrases associated with intent, particularly those that indicate a desire or action to do something

No Comments

No Known Activations

phrases associated with intent, particularly those that indicate a desire or action to do something

No Comments

No Known Activations