INDEX

Explanations

references to URLs and web links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

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

存于互联网档案馆

-0.71

zepine

-0.60

 eventdata

-0.60

moveToFirst

-0.57

newtheorem

-0.56

 betweenstory

-0.55

endregion

-0.54

ImageContext

-0.54

HasKey

-0.53

opardi

-0.52

POSITIVE LOGITS

url

1.43

urls

1.28

Url

1.17

URL

1.16

 urls

1.10

url

1.00

Urls

0.96

URL

0.94

 URLs

0.91

Url

0.89

Activations Density 0.252%

references to URLs and web links

No Comments

No Known Activations