INDEX

Explanations

topics related to social values and structures

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

KommentareTeilen

-0.83

 nahilalakip

-0.73

يميديا

-0.73

GOTREF

-0.71

ModelSerializer

-0.71

JvmStatic

-0.67

UserScript

-0.67

----</

-0.66

)_/¯

-0.66

 createSprite

-0.65

POSITIVE LOGITS

 earlier

0.27

 last

0.26

 past

0.26

his

0.25

 museos

0.25

sky

0.25

 colleagues

0.24

mbggenerated

0.24

sā

0.23

 вра

0.23

Activations Density 0.125%

topics related to social values and structures

No Comments

No Known Activations