INDEX

Explanations

numeric values, particularly in relation to mathematical approximations and scientific notation

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

 celles

-0.42

 esclavos

-0.42

 olun

-0.42

dotta

-0.41

 AssemblyCompany

-0.40

 Überras

-0.39

Cuáles

-0.38

 powin

-0.37

 sejam

-0.37

écart

-0.37

POSITIVE LOGITS

 EconPapers

0.69

 kasarigan

0.68

Espèce

0.62

Personensuche

0.59

 informée

0.57

parsedMessage

0.57

 autorytatywna

0.56

Ӕ

0.56

 nakalista

0.56

 odonto

0.55

Activations Density 0.005%

numeric values, particularly in relation to mathematical approximations and scientific notation

No Comments

No Known Activations