INDEX

Explanations

negative associations and relationships in data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

up

-0.72

mobileqq

-0.60

 doInBackground

-0.58

WireFormatLite

-0.57

исправ

-0.56

">+

-0.56

ołnier

-0.56

multirow

-0.55

}_{+

-0.54

Up

-0.54

POSITIVE LOGITS

(-)

1.07

InjectAttribute

0.86

(-)

0.79

DropColumn

0.73

=-\

0.70

FailureListener

0.69

$(-

0.68

 violets

0.66

 Tampoco

0.66

NullException

0.66

Activations Density 0.663%

negative associations and relationships in data

No Comments

No Known Activations