INDEX

Explanations

references to figures or tables in a document

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

markdown link anchors

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

UserScript

-0.56

GEBURTSDATUM

-0.55

AddTagHelper

-0.51

awtextra

-0.50

 financières

-0.50

 humaines

-0.50

حياته

-0.48

XMLSchema

-0.48

transQ

-0.47

MemoryWarning

-0.46

POSITIVE LOGITS

}[

0.65

}[!

0.47

 defaultstate

0.46

Abitanti

0.46

 AssemblyCulture

0.44

itsub

0.43

_{[

0.42

хьтан

0.40

[^

0.39

$[

0.39

Activations Density 0.001%

references to figures or tables in a document

markdown link anchors

No Comments

No Known Activations