INDEX

Explanations

references to user profiles or error messages related to profiles

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

 Geschäfte

-0.42

 Ejército

-0.37

 ejército

-0.36

 universidad

-0.36

 griego

-0.35

Exemples

-0.35

Источники

-0.34

 refugiados

-0.34

 catalana

-0.34

 Naciones

-0.34

POSITIVE LOGITS

TagMode

0.60

 gale

0.58

 InputDecoration

0.57

AndEndTag

0.55

Datuak

0.55

StoryboardSegue

0.54

aarrggbb

0.54

apimachinery

0.52

 BoxDecoration

0.51

 BoxFit

0.51

Activations Density 0.152%

references to user profiles or error messages related to profiles

No Comments

No Known Activations

references to user profiles or error messages related to profiles

No Comments

No Known Activations