INDEX

Explanations

negations and phrases expressing opposition or inconsistency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RegistryLite

-0.47

 properly

-0.44

VALIDATE

-0.44

nein

-0.41

 naturel

-0.40

won

-0.40

jog

-0.40

Lup

-0.39

 suitably

-0.39

isNotBlank

-0.39

POSITIVE LOGITS

 estekak

0.94

 Normdatei

0.91

 BoxFit

0.82

 مشين

0.81

tagext

0.78

 consultato

0.75

lccn

0.73

 ujednoznacz

0.70

 GenerationType

0.68

évaluateur

0.67

Activations Density 0.412%

negations and phrases expressing opposition or inconsistency

No Comments

No Known Activations