INDEX

Explanations

references to groups, individuals, or situations that involve multiple entities or contexts

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

specifying conditions or scope

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

-0.33

 Explained

-0.30

 déf

-0.29

 poli

-0.29

 Poli

-0.29

 Among

-0.29

 préf

-0.29

 Dead

-0.29

Und

-0.28

POSITIVE LOGITS

 Numerade

0.83

 ویکی‌پدی

0.81

 Administrativna

0.76

تقاوى

0.76

 Paglinawan

0.75

matchCondition

0.74

Cyfarwyddwr

0.73

oa̍t

0.71

Personendaten

0.69

 EconPapers

0.67

Activations Density 0.039%

references to groups, individuals, or situations that involve multiple entities or contexts

specifying conditions or scope

No Comments

No Known Activations