INDEX

Explanations

names and titles associated with specific individuals and organizations

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

names of people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

ParallelGroup

-0.53

])))

-0.50

Pow

-0.47

odacty

-0.46

 Norton

-0.46

 withdrawals

-0.45

โก

-0.44

qrt

-0.43

Cas

-0.43

or

-0.42

POSITIVE LOGITS

 spearheaded

0.71

 sendiri

0.67

帖最后由

0.65

 partenariat

0.63

 samarbeid

0.63

 partnered

0.63

 CoreData

0.62

 herself

0.60

 opérés

0.60

 teamed

0.59

Activations Density 0.121%

names and titles associated with specific individuals and organizations

names of people and places

No Comments

No Known Activations