INDEX

Explanations

phrases related to complex studies or analyses involving multiple components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 AttributeSet

-0.80

homonymie

-0.63

 comprando

-0.62

 InputDecoration

-0.62

Geplaatst

-0.58

 worshipped

-0.56

 altars

-0.54

 diffusivity

-0.54

 spirituale

-0.54

WarningLevel

-0.54

POSITIVE LOGITS

 lengthy

0.86

 brief

0.83

 extensive

0.77

 successful

0.74

 conducted

0.70

essive

0.69

 protracted

0.69

多次

0.67

successful

0.66

 unsuccessful

0.66

Activations Density 0.636%

phrases related to complex studies or analyses involving multiple components

No Comments

No Known Activations