INDEX

Explanations

phrases indicating background context or introductory information

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

transition marker

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 informée

-0.45

tagHelperRunner

-0.35

 BufferedWriter

-0.35

فایل‌لار

-0.35

'])){

-0.35

__":

-0.34

 nonUne

-0.34

MockMvc

-0.34

доми

-0.33

TagMode

-0.33

POSITIVE LOGITS

 الرياضيه

0.55

featureID

0.49

AntiForgeryToken

0.47

 digress

0.47

WriteTagHelper

0.46

 linkovi

0.45

 Colo

0.44

 ligiloj

0.44

ushchev

0.44

BufferException

0.43

Activations Density 0.040%

phrases indicating background context or introductory information

transition marker

No Comments

No Known Activations