INDEX

Explanations

mathematical equations and notations related to subsets and their properties

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

Following the word "in"

np_acts-logits-general · gemini-2.0-flash

membership or non-membership in lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 desconoc

-0.40

PostExecute

-0.39

ÑOS

-0.38

 propOrder

-0.36

 decía

-0.35

 Wikiseite

-0.34

 Vortrag

-0.34

erschied

-0.33

 LIRE

-0.33

Спасылкі

-0.32

POSITIVE LOGITS

 list

0.64

 shortlist

0.54

IsContent

0.52



0.52

 lista

0.51

 daftar

0.51

 liste

0.50

achel

0.48

migrationBuilder

0.48

∉

0.48

Activations Density 0.602%

mathematical equations and notations related to subsets and their properties

Following the word "in"

membership or non-membership in lists

No Comments

No Known Activations