INDEX

Explanations

instances of object creation and manipulation in a programming context

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

new object instantiation

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

 للمعارف

-0.75

 autorytatywna

-0.73

 ivelany

-0.67

 betweenstory

-0.66

AISSEE

-0.66

tvguidetime

-0.64

 estekak

-0.63

 myſelf

-0.63

 мәкал

-0.61

 faſt

-0.61

POSITIVE LOGITS

0.50

0.48

can

0.48

is

0.46

 only

0.44

has

0.44

 also

0.43

 einge

0.43

 used

0.42

usercontent

0.41

Activations Density 0.044%

instances of object creation and manipulation in a programming context

new object instantiation

No Comments

No Known Activations