INDEX

Explanations

references to collective entities or groups of people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 starb

-0.87

 Drapeau

-0.76

مقاله

-0.68

ينما

-0.67

 avoient

-0.67

 Jérusalem

-0.66

 proximité

-0.66

Diwedd

-0.63

 auroit

-0.63

 beginnetje

-0.62

POSITIVE LOGITS

 lotta

1.01

lot

0.90

big

0.89

 lots

0.85

 mucha

0.84

Mucha

0.75

Lots

0.75

 some

0.75

 guys

0.74

lots

0.73

Activations Density 0.078%

references to collective entities or groups of people

No Comments

No Known Activations

references to collective entities or groups of people

No Comments

No Known Activations