INDEX

Explanations

references to notable individuals and their expertise or influence in various fields

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

leaders and experts

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

 Potential

-0.39

Potential

-0.39

ras

-0.39

awtextra

-0.37

SPATH

-0.36

 Half

-0.36

 архивлан

-0.36

MIDDLEWARE

-0.36

 کہ

-0.36

WindowConstants

-0.35

POSITIVE LOGITS

 expert

0.69

expert

0.67

 EXPERT

0.59

 AssemblyCompany

0.55

 expertise

0.54

 experts

0.53

eseorang

0.51

 expertos

0.50

 experto

0.50

XMLSchema

0.50

Activations Density 0.014%

references to notable individuals and their expertise or influence in various fields

leaders and experts

No Comments

No Known Activations