INDEX

Explanations

references to tourists

oai_token-act-pair · gpt-3.5-turbo

references to tourists and tourism-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rity

-0.73

kus

-0.71

ntax

-0.66

avez

-0.65

ris

-0.65

-0.63

rah

-0.63

efer

-0.63

 equations

-0.63

à¼

-0.63

POSITIVE LOGITS

 attractions

1.04

 tourist

1.03

Tour

0.96

 attraction

0.93

 tourism

0.91

 tourists

0.84

ESCO

0.78

 Tourism

0.78

 visa

0.76

 abroad

0.75

Activations Density 0.009%

references to tourists

references to tourists and tourism-related activities

No Comments

No Known Activations