INDEX

Explanations

words related to specific personal names

oai_token-act-pair · gpt-3.5-turbo

variations of the prefix "An" often associated with names or titles

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

glers

-0.62

 cones

-0.62

FIG

-0.61

 fork

-0.61

 caution

-0.61

 Typhoon

-0.59

 USSR

-0.58

 hail

-0.58

 redundancy

-0.57

 Cron

-0.57

POSITIVE LOGITS

wered

1.05

imity

0.96

illary

0.96

heim

0.94

otropic

0.91

obic

0.85

azi

0.84

omaly

0.83

asia

0.83

azon

0.82

Activations Density 0.066%

words related to specific personal names

variations of the prefix "An" often associated with names or titles

No Known Activations