INDEX

Explanations

references to specific people, places, or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

adaptiveStyles

-0.60

Panamoan

-0.44

GenerationType

-0.42

StoryboardSegue

-0.41

 محفوظة

-0.41

 Daun

-0.41

 Италијани

-0.41

 GenerationType

-0.40

[]:

-0.40

webElementXpaths

-0.40

POSITIVE LOGITS

出版年

0.48

jojo

0.40

Bab

0.39

 AssemblyCulture

0.37

下载附件

0.37

Jojo

0.37

muñeca

0.36

ૢ

0.36

 dada

0.36

Bab

0.36

Activations Density 0.555%

references to specific people, places, or entities

No Comments

No Known Activations