INDEX

Explanations

terms that indicate importance or effectiveness in various contexts

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UnknownFields

-0.56

igshid

-0.53

Hochspringen

-0.49

 IndexPath

-0.49

tahui

-0.47

 Officers

-0.47

withId

-0.47

ั้ง

-0.46

ın

-0.46

ync

-0.46

POSITIVE LOGITS

 itſelf

0.87

^(@)

0.81

styleType

0.81

 TimeUnit

0.81

$_"

0.79

neſs

0.75

 ویکی‌پدیای

0.73

 onderdeel

0.71

PerformLayout

0.67

 themſelves

0.67

Activations Density 0.019%

terms that indicate importance or effectiveness in various contexts

No Comments

No Known Activations