INDEX

Explanations

technical terminology and definitions

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

supra

-0.50

RTEX

-0.49

LLocation

-0.48

ifikationer

-0.47

LEncoder

-0.47

Cyfeiriadau

-0.47

GRANTED

-0.45

Хьажоргаш

-0.45

 Kela

-0.45

addPreferredGap

-0.44

POSITIVE LOGITS

 <=",

0.43

ฬ

0.42

yaç

0.38

DispatchToProps

0.37

siębior

0.37

Cyfarwyddwr

0.36

KURZBESCHREIBUNG

0.34

lück

0.33

fromnode

0.33

mäß

0.33

Activations Density 0.101%

technical terminology and definitions

No Comments

No Known Activations