INDEX

Explanations

words related to specific brands and products, particularly in technology or computing contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

 Tatsache

-0.40

 modelAndView

-0.35

 caída

-0.30

 fact

-0.29

 Pingback

-0.29

ivamente

-0.28

Antal

-0.28

 omge

-0.28

 Wirklichkeit

-0.28

localctx

-0.28

POSITIVE LOGITS

Che

0.91

CHE

0.83

Che

0.79

che

0.76

Cheat

0.69

 Cheat

0.68

 Chee

0.68

 Ches

0.66

 Chel

0.66

 Chew

0.66

Activations Density 0.156%

words related to specific brands and products, particularly in technology or computing contexts

No Comments

No Known Activations