INDEX

Explanations

phrases emphasizing possession and personal references related to users and entities

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

following "the"

np_acts-logits-general · gemini-2.0-flash

the following entity

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 &___

-0.64

 שוליים

-0.60

 محفوظة

-0.60

pozdrawiam

-0.59

andinav

-0.58

 propOrder

-0.56

上角

-0.56

'][]

-0.56

 []).

-0.55

 ($("#

-0.53

POSITIVE LOGITS

webElementXpaths

0.68

setVerticalGroup

0.65

 betweenstory

0.63

mybatisplus

0.62

UrlResolution

0.53

Datuak

0.53

 CreateTagHelper

0.52

 samfun

0.52

 própri

0.51

 InputDecoration

0.51

Activations Density 0.551%

phrases emphasizing possession and personal references related to users and entities

following "the"

the following entity

No Comments

No Known Activations