INDEX

Explanations

related character traits and attributes

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

describes character states and actions

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

 yourselves

-0.39

Denomin

-0.38

 Италијани

-0.38

 Sult

-0.36

ilage

-0.36

dafx

-0.36

summer

-0.35

 franqu

-0.35

 calibration

-0.34

stimulus

-0.33

POSITIVE LOGITS

GEBURTSDATUM

0.63

Personendaten

0.56

 surla

0.50

WithMany

0.50

antMatchers

0.49

0.48

ViewFeatures

0.48

setof

0.47

 FetchType

0.46

sidemargin

0.45

Activations Density 0.063%

related character traits and attributes

describes character states and actions

No Comments

No Known Activations