INDEX

Explanations

technical terms and keywords related to programming and system processes

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

Fra

-0.39

fra

-0.31

<eos>

-0.30

SuppressLint

-0.28

 following

-0.27

Fra

-0.26

Sch

-0.26

 which

-0.26

 assert

-0.25

__;

-0.24

POSITIVE LOGITS

ChildScrollView

1.13

 препратки

0.97

دانشنامهٔ

0.89

Diwedd

0.87

Personensuche

0.85

 defaultstate

0.82

تقاوى

0.81

uxxxx

0.79

 AssemblyCulture

0.78

 للاسماء

0.77

Activations Density 0.004%

technical terms and keywords related to programming and system processes

No Comments

No Known Activations