INDEX

Explanations

instances of the word "help" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 nowhere

-0.58

 său

-0.56

 tolerated

-0.56

 للمعارف

-0.53

ยน

-0.52

Hardware

-0.52

 préférable

-0.51

 creș

-0.50

oded

-0.50

гро

-0.50

POSITIVE LOGITS

ItemIndex

0.53

หวัด

0.52

matchCondition

0.50

ArgsConstructor

0.50

 defaultstate

0.50

Descriere

0.49

 octavia

0.48

талья

0.48

HasMaxLength

0.48

 Италијани

0.47

Activations Density 0.272%

instances of the word "help" in various contexts

No Comments

No Known Activations