INDEX

Explanations

numerical data and statistics related to trends and differences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

 kasarigan

-0.70

новништво

-0.58

ImageContext

-0.57

 Louisa

-0.56

 للمعارف

-0.56

 FontFamily

-0.55

 surla

-0.54

rungsseite

-0.54

 TextInputType

-0.54

 autorytatywna

-0.54

POSITIVE LOGITS

ambién

0.50

 nevoie

0.40

 varandra

0.39

 sebagainya

0.39

 keduanya

0.38

 diantaranya

0.36

 sekitarnya

0.36

occasione

0.35

 jäsen

0.34

 weiteren

0.34

Activations Density 0.619%

numerical data and statistics related to trends and differences

No Comments

No Known Activations