INDEX

Explanations

fragments of personal identity and self-referential expressions

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

my own family experiences

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

AndroidJUnit

-0.54

writeField

-0.52

styleType

-0.51

SerializedSize

-0.48

 Chwiliwch

-0.46

prefixer

-0.46

WireFormat

-0.45

PyExc

-0.44

nolia

-0.43

SystemColors

-0.42

POSITIVE LOGITS

CDCl

0.38

ensas

0.37

rotating

0.37

iented

0.34

ETHING

0.34

 قهر

0.34

exclusive

0.34

aéroport

0.33

تقاوى

0.33

 lebaran

0.33

Activations Density 0.010%

fragments of personal identity and self-referential expressions

my own family experiences

No Comments

No Known Activations