INDEX

Explanations

patterns related to data analysis and modeling metrics

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

percentages and ratios

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

 InputDecoration

-0.54

Rujuakan

-0.54

تفصیلات

-0.52

richTextPanel

-0.50

UserScript

-0.50

 समीक्षक

-0.49

&)

-0.49

Hozzáférés

-0.49

 esternos

-0.48

󠁿

-0.47

POSITIVE LOGITS

 Keuangan

0.35

long

0.34

transQ

0.33

WithIdentifier

0.33

 alkal

0.33

あらすじ

0.32

แก่

0.32

 circonstances

0.32

///</

0.31

instancetype

0.31

Activations Density 0.007%

patterns related to data analysis and modeling metrics

percentages and ratios

No Comments

No Known Activations