INDEX

Explanations

references to entities involved in agreements or negotiations

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 resourceCulture

-0.38

//

-0.34

pie

-0.33

 threatening

-0.32

 Utilizamos

-0.29

 savory

-0.29

Color

-0.29

vz

-0.28

 color

-0.28

Tpl

-0.28

POSITIVE LOGITS

AntiForgeryToken

0.67

存于互联网档案馆

0.56

 queſta

0.55

ActionCreators

0.55

ksikon

0.54

 ProtoMessage

0.54

 OMITBAD

0.53

RegressionTest

0.53

ynchronously

0.52

 &___

0.51

Activations Density 0.034%

references to entities involved in agreements or negotiations

No Comments

No Known Activations