INDEX

Explanations

structural elements and markers within written text

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 AssemblyCulture

-0.59

 PeEnEo

-0.47

ISupport

-0.43

gbaar

-0.42

fromnode

-0.40

Behaviors

-0.40

afs

-0.40

 Infórmanos

-0.40

AnchorStyles

-0.39

ViewImports

-0.38

POSITIVE LOGITS

ContentLoaded

0.47

erçe

0.46

 linkovi

0.39

transient

0.37

 Préférences

0.37

 snatched

0.37

 corporativa

0.36

 csal

0.36

EndGlobalSection

0.36

læ

0.35

Activations Density 0.007%

structural elements and markers within written text

No Comments

No Known Activations