INDEX

Explanations

punctuation marks and symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

 متعلقه

-1.08

 Roskov

-0.80

tură

-0.78

fromnode

-0.76

\{\\

-0.73

)"),

-0.73

úgó

-0.71

ftagPool

-0.70

ainville

-0.69

سطس

-0.68

POSITIVE LOGITS

页面存档备份

0.43

 egiten

0.42

Gweler

0.39

 Wikimedijinoj

0.38

bi

0.38

../

0.37

 pran

0.36

قایناقلار

0.36

 resourceCulture

0.35

 Chon

0.35

Activations Density 0.461%

punctuation marks and symbols

No Comments

No Known Activations