INDEX

Explanations

phrases and values related to numerical ranges and limits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 nano

-0.40

テンツ

-0.36

RTHOOK

-0.35

mod

-0.35

 Horizontal

-0.34

 poul

-0.34

 minor

-0.33

 التن

-0.33

Piz

-0.33

ca

-0.32

POSITIVE LOGITS

MemoryWarning

0.80

istoitu

0.65

ConstraintMaker

0.60

SharedDtor

0.60

Personendaten

0.59

fjspx

0.59

EDEFAULT

0.58

BeginInit

0.57

 gawas

0.57

 vacacionales

0.56

Activations Density 0.534%

phrases and values related to numerical ranges and limits

No Comments

No Known Activations