INDEX

Explanations

instances of the word "using."

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

how something is done

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

 pleaſure

-0.63

 itſelf

-0.62

 themſelves

-0.60

 perſon

-0.57

ſelf

-0.57

 houſe

-0.55

 himſelf

-0.55

ſelves

-0.54

 cauſe

-0.52

 laſt

-0.50

POSITIVE LOGITS

 utilising

0.83

 using

0.78

 utilizing

0.76

 menggunakan

0.73

 utilizzando

0.72

 utilizando

0.72

__(/*!

0.69

 mittels

0.69

 USING

0.68

Using

0.66

Activations Density 0.054%

instances of the word "using."

how something is done

No Comments

No Known Activations