INDEX

Explanations

references to specific locations or notable buildings

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

geographical and common phrases

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

IntoConstraints

-0.46

 nahilalakip

-0.41

 else

-0.40

 demografica

-0.38

 static

-0.34

 sentence

-0.34

Népesség

-0.34

 score

-0.34

مصادر

-0.33

 nakalista

-0.33

POSITIVE LOGITS

 Autorizaciones

0.51

Becoming

0.50

 dragen

0.50

 bereiken

0.50

Извест

0.49

faßt

0.48

 oorsp

0.47

AsUp

0.45

Offic

0.45

favorited

0.45

Activations Density 0.040%

references to specific locations or notable buildings

geographical and common phrases

No Comments

No Known Activations