INDEX

Explanations

references to geographic locations, specifically those related to Saint or similar named entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

enumi

-0.82

 propOrder

-0.78

 ClassNotFound

-0.74

Autoritní

-0.64

UserScript

-0.62

 Dimen

-0.60

nelse

-0.59

ällningar

-0.58

ecology

-0.57

jface

-0.57

POSITIVE LOGITS

 Saint

1.05

Saint

1.00

St

0.91

 Sankt

0.89

 saint

0.87

 SAINT

0.86

ankt

0.85

SAINT

0.75

St

0.74

saint

0.71

Activations Density 0.086%

references to geographic locations, specifically those related to Saint or similar named entities

No Comments

No Known Activations