INDEX

Explanations

references to governmental departments and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 esternos

-0.67

 betweenstory

-0.63

PostInfinity

-0.60

 Wikimedijinoj

-0.58

Jîn

-0.53

 culturelle

-0.52

ISupport

-0.49

 houſe

-0.49

intios

-0.48

 grève

-0.48

POSITIVE LOGITS

 Department

0.55

Department

0.46

 Departments

0.42

 organisations

0.42

 department

0.41

geführt

0.40

 organisation

0.39

 DEPARTMENT

0.38

Dept

0.38

 xmlns

0.38

Activations Density 0.004%

references to governmental departments and organizations

No Comments

No Known Activations