INDEX

Explanations

positional indicators in a spatial context

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aarrggbb

-0.74

IonicModule

-0.58

 faſt

-0.56

leſs

-0.54

 BorderSide

-0.53

 queſta

-0.53

tinyos

-0.53

 autorytatywna

-0.52

ſelf

-0.51

 kasarigan

-0.51

POSITIVE LOGITS

 AssemblyCompany

0.51

 calculées

0.34

 refer

0.33

семь

0.32

obiety

0.32

 Hälfte

0.32

 فريبيس

0.31

 depreciation

0.31

InSection

0.30

newtheorem

0.30

Activations Density 0.000%

positional indicators in a spatial context

No Comments

No Known Activations