INDEX

Explanations

concepts related to properties and effects in theoretical discussions

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

Followed by a verb ("is", "was", "can", etc.)

np_acts-logits-general · gemini-2.0-flash

abstract concepts and observations

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

脚注の使い方

-0.57

verwijspagina

-0.55

 ſind

-0.55

adaptiveStyles

-0.52

":[{

-0.51

ellido

-0.51

 propOrder

-0.50

PerformLayout

-0.50

featureID

-0.49

 esternos

-0.49

POSITIVE LOGITS

 platí

0.43

 heridos

0.43

@"/

0.40

 also

0.40

 transmettre

0.40

这点

0.38

 salvación

0.37

 mismos

0.37

 âgé

0.37

 generaciones

0.36

Activations Density 0.455%

concepts related to properties and effects in theoretical discussions

Followed by a verb ("is", "was", "can", etc.)

abstract concepts and observations

No Comments

No Known Activations