INDEX

Explanations

placeholder tokens or indicators of conversation structure and formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 endwhile

-0.57

gige

-0.55

曖昧さ回避

-0.53

ChildScrollView

-0.53

󠁣

-0.53

ngdoc

-0.52

NDL

-0.52

タジー

-0.51

amerikan

-0.50

▪

-0.50

POSITIVE LOGITS

0.85

expandindo

0.80

 الاطلاع

0.77

Билгалдахарш

0.75

0.70

ResumeLayout

0.69

">//

0.67

 Roskov

0.66

#+#

0.64

SourceChecksum

0.62

Activations Density 0.367%

placeholder tokens or indicators of conversation structure and formatting

No Comments

No Known Activations