INDEX

Explanations

references to legal petitions and their associated terms

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

preceding "er" suffix

np_acts-logits-general · gemini-2.0-flash

petition and prisoner

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

MENAFN

-0.72

writeFieldEnd

-0.58

ISupport

-0.56

basicConfig

-0.55

DeleteBehavior

-0.52

tvguidetime

-0.51

contentLoaded

-0.49

fjspx

-0.49

 noDo

-0.49

 EClass

-0.48

POSITIVE LOGITS

er

2.36

ers

2.12

ER

1.91

ERS

1.65

er

1.37

erin

1.27

Er

1.27

erer

1.27

erst

1.27

ered

1.26

Activations Density 0.695%

references to legal petitions and their associated terms

preceding "er" suffix

petition and prisoner

No Comments

No Known Activations