INDEX

Explanations

elements of text that indicate data retrieval or response

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

はじめに

-0.36

canestro

-0.33

 wrestler

-0.32

 plie

-0.32

 indefin

-0.32

 synag

-0.31

 wrestlers

-0.31

 Ballet

-0.31

 agon

-0.30

 Lionel

-0.30

POSITIVE LOGITS

featureID

0.55

EDEFAULT

0.53

ValueGeneration

0.53

berdayakan

0.52

 scheinen

0.50

FetchType

0.50

 seemingly

0.48

Espèce

0.48

ücksich

0.48

 ModelExpression

0.48

Activations Density 0.805%

elements of text that indicate data retrieval or response

No Comments

No Known Activations