INDEX

Explanations

connections and relationships between concepts or items

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

mesis

-0.47

 CascadeType

-0.46

joba

-0.46

 Desenvolvimento

-0.46

koy

-0.45

hcm

-0.45

 potenciales

-0.45

føl

-0.44

SerializeField

-0.44

waltung

-0.44

POSITIVE LOGITS

 highly

0.56

 CreateTagHelper

0.52

featureID

0.51

 extremely

0.47

 épaules

0.46

 ivelany

0.46

的で

0.46

 economically

0.45

 efficient

0.44

しかも

0.44

Activations Density 0.307%

connections and relationships between concepts or items

No Comments

No Known Activations