INDEX

Explanations

words related to user needs and functionalities of products or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

 Biôgrafia

-0.79

 Савезне

-0.78

 disambiguazione

-0.75

исленность

-0.73

 يتيمه

-0.67

NameInMap

-0.66

 Wikimedijinoj

-0.64

ArrowToggle

-0.61



-0.60

 intptr

-0.59

POSITIVE LOGITS

 unparalleled

0.93

 exceptional

0.84

 excellent

0.83

 effortless

0.77

 unsurpassed

0.75

 easy

0.74

 endless

0.73

 reliable

0.72

 unmatched

0.71

 boundless

0.71

Activations Density 0.364%

words related to user needs and functionalities of products or services

No Comments

No Known Activations