INDEX

Explanations

phrases and constructs related to identity and roles

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 postIndex

-0.60

هد

-0.58

 INVENTION

-0.54

小姐

-0.54

 karte

-0.53

olerance

-0.51

 OMIT

-0.51

нор

-0.50

 crudo

-0.49

 kıs

-0.49

POSITIVE LOGITS

Portale

0.89

 houſe

0.76

paravant

0.74

ργο

0.71

 ſch

0.70

etheless

0.69

endphp

0.69

 practicing

0.68

 пожалуйста

0.68

endpush

0.67

Activations Density 0.228%

phrases and constructs related to identity and roles

No Comments

No Known Activations