INDEX

Explanations

sequences of numbers and terms associated with data and measurements

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

numbers and symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 katholischen

-0.46

ⓧ

-0.44

 sebagaimana

-0.43

Див

-0.42

 aynı

-0.41

 mismos

-0.41

 mêmes

-0.40

 Alfaro

-0.40

 bruja

-0.40

 Başkanı

-0.40

POSITIVE LOGITS

 myſelf

0.54

exitRule

0.51

 Autorisations

0.50

RenderAtEndOf

0.49

 esternos

0.48

$_"

0.48

WithIOException

0.48

 betweenstory

0.47

⟬

0.46

 fevere

0.46

Activations Density 0.595%

sequences of numbers and terms associated with data and measurements

numbers and symbols

No Comments

No Known Activations