INDEX

Explanations

phrases related to announcements and results

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

announcements, results, and revelations

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

 esternos

-0.57

tagext

-0.56

Šaltiniai

-0.53

 الرياضيه

-0.52

ViewFeatures

-0.51

 arşivlendi

-0.50

 يتيمه

-0.48

 tartalomajánló

-0.47

OGND

-0.46

 препратки

-0.46

POSITIVE LOGITS

 announced

0.42

 suspense

0.41

 announcement

0.40

reveal

0.40

 announcements

0.40

 объ

0.39

 reveal

0.39

BufferException

0.39

umumkan

0.38

Reve

0.38

Activations Density 0.012%

phrases related to announcements and results

announcements, results, and revelations

No Comments

No Known Activations