INDEX

Explanations

scientific terminology related to measurement or analysis

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

 Numerade

-0.60

 ſta

-0.58

 Италијани

-0.57

dji

-0.56

protoimpl

-0.54

AnimationsModule

-0.52

 nonUne

-0.52

GTCX

-0.50

 ſte

-0.48

ſp

-0.48

POSITIVE LOGITS

 useAppContext

0.41

 getApplication

0.36

 endforeach

0.36

 according

0.34

porre

0.33

tahankan

0.33

 liệt

0.32

eachtet

0.32

 volgens

0.32

Хьажоргаш

0.30

Activations Density 0.194%

scientific terminology related to measurement or analysis

No Comments

No Known Activations