INDEX

Explanations

negative or low numeric values, particularly in relation to certain parameters or terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

 Warne

-0.60

reign

-0.48

 Munt

-0.45

AddRange

-0.43

 Gains

-0.42

alış

-0.42

朋

-0.42

 walls

-0.42

straw

-0.42

écri

-0.42

POSITIVE LOGITS

NameInMap

0.85

 pinulongan

0.83

 autorytatywna

0.79

 Réponses

0.79

AsUp

0.76

ModelAdmin

0.75

RefNanny

0.74

+#+#

0.69

 cherchés

0.69

 дописавши

0.68

Activations Density 0.437%

negative or low numeric values, particularly in relation to certain parameters or terms

No Comments

No Known Activations