INDEX

Explanations

mathematical symbols and expressions related to formal definitions or equations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ValueStyle

-0.40

 crossorigin

-0.40

❋

-0.36

綜合報導

-0.35

ơn

-0.33

 equalTo

-0.32

 fucking

-0.31

대한

-0.31

 prakty

-0.31

IsPostBack

-0.30

POSITIVE LOGITS

parsedMessage

0.75

OGND

0.72

SourceChecksum

0.61

MessageOf

0.60

+#+#

0.58

 fieldNum

0.52

 oprot

0.52

 intStringLen

0.51

 betweenstory

0.51

featureID

0.51

Activations Density 14.390%

mathematical symbols and expressions related to formal definitions or equations

No Comments

No Known Activations