INDEX

Explanations

proper nouns, particularly names and geographic locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Muske

-1.01

KommentareTeilen

-0.93

 doigts

-0.93

 Cheshire

-0.85

ัจ

-0.84

+:+

-0.83

Ches

-0.82

 Dumas

-0.82

 Púb

-0.78

 ogrodow

-0.78

POSITIVE LOGITS

 "..\..\

0.86

 propOrder

0.81

GEBURTSDATUM

0.78

Rode

0.78

 Rode

0.74

intios

0.71

 חיצוניים

0.69

XmlAccessorType

0.68

Teh

0.67

 Chalmers

0.67

Activations Density 2.714%

proper nouns, particularly names and geographic locations

No Comments

No Known Activations