INDEX

Explanations

numbers or symbols indicating repetition or multiplication

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

mathematical multiplication operations

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

HomeAsUpEnabled

-0.58

AllowUser

-0.51

 indígen

-0.49

.」

-0.48

 Anſ

-0.46

 diken

-0.46

adaptiveStyles

-0.46

tangentMode

-0.45

 santuario

-0.45

 löytyy

-0.44

POSITIVE LOGITS

0.91

0.79

()*

0.77

*=

0.73

*(

0.63

$*$

0.63

×

0.63

(*

0.62

*$

0.62

times

0.61

Activations Density 0.208%

numbers or symbols indicating repetition or multiplication

mathematical multiplication operations

No Comments

No Known Activations