INDEX

Explanations

references to the prefix "de-" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 виправивши

-0.99

WebVitals

-0.87

 AttributeSet

-0.82

 Roskov

-0.81

complexContent

-0.77

RegistryLite

-0.76

Спасылкі

-0.76

 препратки

-0.75

UnsafeEnabled

-0.75

 createState

-0.75

POSITIVE LOGITS

 Diſ

0.64

jenigen

0.56

 Reſ

0.56

selben

0.55

 Ural

0.54

 nying

0.53

enumi

0.52

DHT

0.52

 Cæsar

0.51

 Dede

0.51

Activations Density 0.456%

references to the prefix "de-" in various contexts

No Comments

No Known Activations