INDEX

Explanations

occurrences of the word "on" in various forms

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

on followed by specific suffixes

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

timedia

-0.57

Fernseh

-0.53

gefügt

-0.52

 fédé

-0.52

aktery

-0.52

 HasFactory

-0.48

racuda

-0.47

fieldType

-0.47

 समीक्षाएं

-0.47

jarati

-0.46

POSITIVE LOGITS

on

1.17

On

1.17

On

1.10

ON

1.09

on

1.02

isOn

0.86

ON

0.84

Ons

0.79

onMouse

0.77

 onto

0.75

Activations Density 0.025%

occurrences of the word "on" in various forms

on followed by specific suffixes

No Comments

No Known Activations