INDEX

Explanations

the repeated use of the word "en" in various contexts

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

prepositions and 'en'

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

 myſelf

-0.73

 pleaſure

-0.72

ſelf

-0.71

 enfans

-0.69

 himſelf

-0.67

 purpoſe

-0.63

 houſe

-0.62

 Monfieur

-0.60

wiſe

-0.60

 itſelf

-0.60

POSITIVE LOGITS

in

0.97

In

0.80

on

0.76

en

0.71

ใน

0.69

In

0.66

On

0.66

IN

0.66

 في

0.65

at

0.63

Activations Density 0.001%

the repeated use of the word "en" in various contexts

prepositions and 'en'

No Comments

No Known Activations