INDEX

Explanations

proper nouns and identifiers in the text

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

we already have

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

contentLoaded

-0.48

featureID

-0.48

TemporalType

-0.47

MemoryWarning

-0.46

CompleteListener

-0.45

IsMutable

-0.45

exitRule

-0.44

 ThemeData

-0.44

 MutableLiveData

-0.43

asantry

-0.43

POSITIVE LOGITS

 nahilalakip

0.42

bkz

0.41

 disambiguazione

0.40

Rè

0.39

期刊论文

0.39

rungsseite

0.38

ఔ

0.35

ぐり

0.35

Urs

0.35

scribe

0.35

Activations Density 0.000%

proper nouns and identifiers in the text

we already have

No Comments

No Known Activations