INDEX

Explanations

mentions of geographical locations specific to India

oai_token-act-pair · gpt-3.5-turbo

geographic references related to Bengal and Bangalore

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mble

-1.12

Ö¼

-1.02

ettings

-0.76

ipple

-0.75

ertodd

-0.68

ynt

-0.67

pez

-0.67

earable

-0.67

OSP

-0.67

bilt

-0.65

POSITIVE LOGITS

uru

1.23

 tigers

0.93

 Bengal

0.87

wings

0.87

icum

0.86

 Nadu

0.80

ensis

0.80

 tiger

0.80

ataka

0.75

avia

0.72

Activations Density 0.006%

mentions of geographical locations specific to India

geographic references related to Bengal and Bangalore

No Comments

No Known Activations