INDEX

Explanations

descriptions of complex interactions or relationships

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

Text following articles or prepositions

np_acts-logits-general · gemini-2.0-flash

book descriptions

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

 dimo

-0.67

 promouvoir

-0.66

!!!)

-0.63

)!

-0.63

 Vordergrund

-0.63

viamente

-0.63

 Spaß

-0.62

 Aufgrund

-0.61

 vœux

-0.61

 vVar

-0.61

POSITIVE LOGITS

__':

0.61

genstein

0.58

__':

0.57

fuck

0.56

 fucked

0.55

 fucking

0.53

sonder

0.52

ppure

0.51

ContentAsync

0.50

 empty

0.49

Activations Density 0.534%

descriptions of complex interactions or relationships

Text following articles or prepositions

book descriptions

No Comments

No Known Activations