INDEX

Explanations

keywords related to technology and healthcare imagery and usage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Билгалдахарш

-1.01

 Paglinawan

-0.74

Бахар

-0.72

mybatisplus

-0.69

UrlResolution

-0.68

awtextra

-0.64

 MenuView

-0.62

PreInfinity

-0.60

 autorytatywna

-0.59

 мәкал

-0.59

POSITIVE LOGITS

 rootReducer

0.73

 itſelf

0.71

 Houſe

0.64

SNA

0.61

 Majefty

0.60

 Theſe

0.59

省市镇

0.58

 Darauf

0.57

كويكب

0.57

Geografie

0.57

Activations Density 0.659%

keywords related to technology and healthcare imagery and usage

No Comments

No Known Activations