INDEX

Explanations

references to creative individuals and their works

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

Non-English text/Unicode characters

np_acts-logits-general · gemini-2.0-flash

non-english words and scripts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Negative Logits

<?

-0.45

 AssemblyCompany

-0.40

opsida

-0.40

ulemon

-0.39

يكب

-0.39

/**

-0.38

letoe

-0.37

ItemBackground

-0.36

__':

-0.36

romyalgia

-0.36

POSITIVE LOGITS

 utafitiHapana

0.70

 rhestr

0.64

 кӀ

0.53

 uska

0.49

Билгалдахарш

0.48

 pinulongan

0.48

RTHOOK

0.47

 BoxFit

0.46

Портали

0.45

lateinit

0.45

Activations Density 0.431%

references to creative individuals and their works

Non-English text/Unicode characters

non-english words and scripts

No Known Activations