INDEX

Explanations

imperative or suggestive phrases that indicate requests or invitations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

favorited

-0.54

 виправивши

-0.53

 rhestr

-0.52

 disambiguazione

-0.52

Demikian

-0.51

agramm

-0.51

 кӀ

-0.51

İstinadlar

-0.49

Havolalar

-0.49

<#

-0.48

POSITIVE LOGITS

us

0.87

 alone

0.82

byg

0.80

 loose

0.78

me

0.78

tuce

0.72

 them

0.68

go

0.66

ResumeLayout

0.66

sizeCache

0.66

Activations Density 0.125%

imperative or suggestive phrases that indicate requests or invitations

No Comments

No Known Activations