INDEX

Explanations

references to sampling or data collection methods

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

track list sections

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

 nahilalakip

-0.68

WaitGroup

-0.63

 الحره

-0.60

 Chwiliwch

-0.59

 gynhyrchwyd

-0.58

zzleHttp

-0.58

 مشين

-0.58

AsUp

-0.57

 varargin

-0.56

tonode

-0.56

POSITIVE LOGITS

DockStyle

0.34

LookAnd

0.33

Cur

0.31

BCD

0.31

fin

0.30

 TType

0.30

Fin

0.29

Con

0.28

Un

0.27

Pas

0.27

Activations Density 0.019%

references to sampling or data collection methods

track list sections

No Comments

No Known Activations