INDEX

Explanations

mathematical notations and statistical terms, particularly related to confidence intervals and variables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

OptionsMenu

-0.38

 estekak

-0.36

 informée

-0.36

 statement

-0.35

󠁣

-0.35

 Administrativna

-0.35

 fotográfica

-0.34

 tắc

-0.34

+#+

-0.33

 morceau

-0.33

POSITIVE LOGITS

 between

0.74

 בין

0.69

Between

0.66

between

0.65

 zwischen

0.65

 Between

0.64

CGRectMake

0.62

 CGRectMake

0.61

 tussen

0.61

 mellem

0.59

Activations Density 0.605%

mathematical notations and statistical terms, particularly related to confidence intervals and variables

No Comments

No Known Activations