INDEX

Explanations

descriptions of food and dining experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 médicale

-0.49

rungsseite

-0.42

 pomeriggio

-0.40

Referencoj

-0.40

 médical

-0.40

 nadzieję

-0.40

transQ

-0.39

 comprimidos

-0.39

 sosial

-0.39

 montant

-0.39

POSITIVE LOGITS

 crunchy

0.53

 flavorful

0.52

 luscious

0.52

SourceChecksum

0.52

 &___

0.50

 chewy

0.49

IsMutable

0.49

sizeCache

0.49

ectady

0.47

 garnish

0.47

Activations Density 0.083%

descriptions of food and dining experiences

No Comments

No Known Activations