INDEX

Explanations

numeric values or identifiers related to product recalls and services

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

1-800 phone numbers

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

principalColumn

-0.42

 InputDecoration

-0.40

 ricev

-0.38

medriver

-0.37

zugehen

-0.36

gegeben

-0.35

 Dinamarca

-0.35

 festen

-0.35

 moderate

-0.35

 rather

-0.34

POSITIVE LOGITS

:✨

0.63

 EconPapers

0.58

 pinulongan

0.54

цездатний

0.54

 Италијани

0.54

évaluateur

0.52

adaptiveStyles

0.52

 CanadaChoose

0.52

ंदीखरीदारी

0.51

Autoritní

0.50

Activations Density 0.004%

numeric values or identifiers related to product recalls and services

1-800 phone numbers

No Comments

No Known Activations