INDEX

Explanations

terms related to indexing or categorizing information

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

indexed by, indexed graph

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

providedIn

-0.55

endphp

-0.44

 CanadaChoose

-0.42

 ThemeData

-0.42

kapnya

-0.38

stdbool

-0.38

ckså

-0.38

 siihen

-0.38

 financiación

-0.37

 kaarangay

-0.37

POSITIVE LOGITS

 indexed

0.82

 Indexed

0.76

indexed

0.71

Indexed

0.63

Personendaten

0.57

 indexing

0.57

 незавершена

0.52

 index

0.51

 indexes

0.50

ništvo

0.47

Activations Density 0.000%

terms related to indexing or categorizing information

indexed by, indexed graph

No Comments

No Known Activations