INDEX

Explanations

phrases related to health metrics and electronic devices

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

housing accommodating

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

MockBean

-0.39

دانشنامهٔ

-0.38

ener

-0.36

 QtCore

-0.36

cie

-0.35

aden

-0.35

ldc

-0.35

Meg

-0.35

 Bill

-0.34

 Mell

-0.34

POSITIVE LOGITS

 ProtoMessage

0.77

 itſelf

0.68

 AssemblyTitle

0.59

 themſelves

0.57

 configureStore

0.56

RTEX

0.55

IsMutable

0.54

 <<<<<<<<<<<<<<

0.53

 myſelf

0.53

 purpoſe

0.52

Activations Density 0.228%

phrases related to health metrics and electronic devices

housing accommodating

No Comments

No Known Activations