INDEX

Explanations

specific identifiers or attributes related to names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

buie

-0.46

 huella

-0.43

 critères

-0.43

AndEndTag

-0.43

 fact

-0.43

 integridad

-0.43

 dabei

-0.42

rvats

-0.42

Indeed

-0.41

 posibilidad

-0.40

POSITIVE LOGITS

BufferException

0.54

 ſta

0.53

 noDo

0.53

 propOrder

0.52

Datuak

0.52

FailureListener

0.52

UserScript

0.51

didSet

0.51

 виправивши

0.51

IntoConstraints

0.50

Activations Density 3.389%

specific identifiers or attributes related to names and titles

No Comments

No Known Activations