INDEX

Explanations

words that contain the letter 'é'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Kev

-0.50

########.

-0.46

SourceChecksum

-0.45

JoinColumn

-0.45

protoimpl

-0.44

 ReactDOM

-0.43

ísimas

-0.43

Enllaces

-0.42

Síguenos

-0.41

 εγ

-0.41

POSITIVE LOGITS

é

0.83

vé

0.72

eacute

0.68

né

0.64

é

0.63

É

0.62

cé

0.58

ē

0.58

hé

0.58

ék

0.57

Activations Density 0.414%

words that contain the letter 'é'

No Comments

No Known Activations