INDEX

Explanations

names and key attributes related to technologies and methodologies

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

reasonable probability

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

uxxxx

-0.80

ConstraintMaker

-0.57

UnusedPrivate

-0.56

.*")]

-0.55

فایل‌لار

-0.52

 فريبيس

-0.52

лтемелер

-0.51

 Photocase

-0.50

 wireType

-0.49

GenerationType

-0.49

POSITIVE LOGITS

 الرياضيه

0.43

_:

0.39

 onFailure

0.38

 getCategory

0.36

 CreateTagHelper

0.35

 OnTrigger

0.34

 byť

0.32

 wikipagina

0.32

まま

0.32

Activations Density 0.217%

names and key attributes related to technologies and methodologies

reasonable probability

No Comments

No Known Activations