INDEX

Explanations

connections between cause and effect in various contexts

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

preceding a period

np_acts-logits-general · gemini-2.0-flash

standalone concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

"):

-0.55

"),

-0.54

outSlope

-0.54

"],

-0.52

出版年

-0.50

"},

-0.50

]")

-0.48

"])

-0.48

 متعلقه

-0.48

__":

-0.48

POSITIVE LOGITS

0.49

RTEX

0.47

InjectAttribute

0.45

UserScript

0.42

 surla

0.40

IsMutable

0.40

 Sugar

0.39

BagLayout

0.38

HIB

0.38

hs

0.37

Activations Density 0.579%

connections between cause and effect in various contexts

preceding a period

standalone concepts

No Comments

No Known Activations