INDEX

Explanations

phrases indicating time and specific periods in history

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

"in" followed by a year or month

np_acts-logits-general · gemini-2.0-flash

in years or months

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

Vidite

-0.45

webElement

-0.41

CppMethod

-0.40

Erstellt

-0.40

ystemet

-0.39

 royaume

-0.39

꼈

-0.35

addComponent

-0.35

Britannique

-0.35

 daño

-0.35

POSITIVE LOGITS

tangentMode

0.51

UnusedPrivate

0.47

 about

0.47

 IndexError

0.45

principalTable

0.45

 Summe

0.44

 Bade

0.44

 ListTile

0.44

AnchorStyles

0.44

estly

0.44

Activations Density 0.345%

phrases indicating time and specific periods in history

"in" followed by a year or month

in years or months

No Comments

No Known Activations