INDEX

Explanations

phrases related to health conditions and their prevalence

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

also been described

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

новништво

-0.59

 المعيارى

-0.57

ArgumentParser

-0.51

titleMargin

-0.48

audiovisuel

-0.47

unteer

-0.45

RepeatedField

-0.45

inafter

-0.44

cipes

-0.44

ifikationer

-0.43

POSITIVE LOGITS

@[+][

0.44

别

0.39

-------------</

0.36

 nelle

0.35

otropy

0.35

別

0.33

IsContent

0.33

 muualla

0.32

timiento

0.32

 vieler

0.31

Activations Density 0.195%

phrases related to health conditions and their prevalence

also been described

No Comments

No Known Activations