INDEX

Explanations

the word "Pres" and its variations, often in contexts related to authorities or official information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 Infórmanos

-0.89

ReusableCell

-0.70

 autorytatywna

-0.69

 Италијани

-0.68

verifyException

-0.66

 Italijanski

-0.66

UrlResolution

-0.65

MessageTagHelper

-0.61

elemField

-0.61

 الرياضيه

-0.60

POSITIVE LOGITS

 Pres

1.15

Pres

1.11

 pres

1.02

 PRES

0.94

pres

0.91

 prescription

0.77

PRES

0.76

 PRESERV

0.69

 Prescription

0.68

 prescribed

0.66

Activations Density 1.215%

the word "Pres" and its variations, often in contexts related to authorities or official information

No Comments

No Known Activations