INDEX

Explanations

organizations or entities based in specific locations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a location or entity's geographical association

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

 tasted

-0.78

bush

-0.72

perty

-0.70

yden

-0.70

worthy

-0.69

 piled

-0.68

ĺħ

-0.68

 challeng

-0.68

 hurled

-0.67

fax

-0.66

POSITIVE LOGITS

 Chic

0.73

clus

0.68

pty

0.63

 Palo

0.62

 Plymouth

0.62

 partnerships

0.62

 partnership

0.61

ahon

0.61

Wan

0.60

 geographically

0.60

Activations Density 0.113%

organizations or entities based in specific locations

phrases indicating a location or entity's geographical association

No Comments

No Known Activations