INDEX

Explanations

geographical names and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 otomatig

-0.53

Rhestr

-0.53

 كومونز

-0.44

 evolution

-0.43

ArrowToggle

-0.43

Personendaten

-0.43

 Evolution

-0.42

Evolution

-0.42

AccessorTable

-0.41

 EVOLUTION

-0.39

POSITIVE LOGITS

LookAnd

0.50

 elsewhere

0.46

 wherever

0.43

NUMX

0.43

 provinces

0.42

Clik

0.41

 élo

0.40

WriteBarrier

0.40

+:+

0.40

hashtags

0.39

Activations Density 0.332%

geographical names and locations

No Comments

No Known Activations