INDEX

Explanations

references to travel destinations

oai_token-act-pair · gpt-3.5-turbo

references to travel-related entities or brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

onial

-0.75

une

-0.70

roman

-0.70

eff

-0.69

elin

-0.68

ulin

-0.66

edi

-0.66

oust

-0.66

forth

-0.65

plex

-0.65

POSITIVE LOGITS

 Travels

1.34

 VIDEOS

1.11

NetMessage

1.04

 Flavoring

1.00

 destinations

0.88

 Tracks

0.83

dayName

0.82

||||

0.82

 Reader

0.79

CHAT

0.78

Activations Density 0.011%

references to travel destinations

references to travel-related entities or brands

No Comments

No Known Activations