INDEX

Explanations

mentions of specific people or characters, particularly variations of the name "N."

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

Follows the letter "N"

np_acts-logits-general · gemini-2.0-flash

words starting with n

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

JAKARTA

-0.57

 contra

-0.50

 Geheim

-0.48

ducation

-0.46

 strict

-0.45

 while

-0.45

otechnology

-0.45

ாம்

-0.44

 bety

-0.44

 dispen

-0.44

POSITIVE LOGITS

ThroughAttribute

0.71

 imageNamed

0.66

ⓧ

0.66

LookAnd

0.64

انيف

0.64

 Nomin

0.63

TestingModule

0.61

 læng

0.60

 onions

0.58

тьяна

0.58

Activations Density 0.264%

mentions of specific people or characters, particularly variations of the name "N."

Follows the letter "N"

words starting with n

No Comments

No Known Activations