INDEX

Explanations

phrases that indicate conditions, importance, or essential actions related to processes or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ValueStyle

-0.52

Efq

-0.49

RetentionPolicy

-0.48

出版年

-0.48

 ویکی‌پدیا

-0.47

بوابة

-0.47

ⓧ

-0.47

ſelf

-0.46

Források

-0.45

 Audiodateien

-0.45

POSITIVE LOGITS

 only

0.58

 very

0.47

 also

0.47

 given

0.45

ఔ

0.43

 however

0.43

only

0.43

 always

0.42

 still

0.41

not

0.39

Activations Density 1.232%

phrases that indicate conditions, importance, or essential actions related to processes or entities

No Comments

No Known Activations