INDEX

Explanations

references to programming or technical components and their properties

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

 Vergangenheit

-0.48

 prosty

-0.44

ioare

-0.43

zaamheid

-0.43

 borbo

-0.43

 Gespräch

-0.43

ificantly

-0.43

 Betrachtung

-0.42

etcode

-0.42

thood

-0.42

POSITIVE LOGITS

 ddelweddau

0.60

WireFormatLite

0.60

getMenuInflater

0.60

 MainAxisSize

0.60

FunctionFlags

0.57

AutoScaleMode

0.54

StructEnd

0.53

 kaarangay

0.53

ScopeManager

0.52

SequentialGroup

0.52

Activations Density 0.317%

references to programming or technical components and their properties

No Comments

No Known Activations