INDEX

Explanations

names and identifiers linked to individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

tagHelperRunner

-0.71

 betweenstory

-0.52

 increí

-0.51

 aanbied

-0.49

momix

-0.49

konos

-0.47

 sabre

-0.47

 zuſammen

-0.47

 creș

-0.47

oulli

-0.46

POSITIVE LOGITS

UW

0.71

 Wisconsin

0.64

UW

0.63

Tai

0.60

Wisconsin

0.58

Tai

0.57

Yi

0.55

 Russell

0.55

 Milwaukee

0.54

uw

0.52

Activations Density 0.330%

names and identifiers linked to individuals or entities

No Comments

No Known Activations