INDEX

Explanations

references to various companies or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

ful

-0.32

Haben

-0.31

SBATCH

-0.31

 TAMBÉM

-0.29

tic

-0.26

coord

-0.26

 бъ

-0.26

fulness

-0.26

 Kokos

-0.26

pad

-0.25

POSITIVE LOGITS

 Administrativna

0.80

 للمعارف

0.75

 برانيه

0.72

 pinulongan

0.72

rungsseite

0.71

 executives

0.71

 CreateTagHelper

0.69

 AssemblyTitle

0.68

bootstrapcdn

0.68

 autorytatywna

0.67

Activations Density 0.103%

references to various companies or organizations

No Comments

No Known Activations