INDEX

Explanations

references to statistical data and numerical information

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

numbers followed by punctuation

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

awtextra

-0.46

 defaultstate

-0.43

 <>",

-0.41

localctx

-0.39

 EconPapers

-0.39

 ostavi

-0.38

 مرئيه

-0.38

RTSC

-0.38

 anún

-0.34

oredCriteria

-0.34

POSITIVE LOGITS

tangentMode

0.52

 betweenstory

0.50

 improvised

0.50

 arşivlendi

0.49

onedDateTime

0.49

ayet

0.47

Wednesday

0.47

tapan

0.46

 linkovi

0.46

 Wicidata

0.45

Activations Density 0.006%

references to statistical data and numerical information

numbers followed by punctuation

No Comments

No Known Activations