INDEX

Explanations

proper names or surnames related to a particular individual named Ben Stein

oai_token-act-pair · gpt-3.5-turbo

the name "Einstein" in various contexts

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

accompan

-0.82

 millenn

-0.80

teenth

-0.78

 runaway

-0.75

profits

-0.74

icable

-0.74

 pestic

-0.72

 tradem

-0.72

phis

-0.70

cled

-0.68

POSITIVE LOGITS

stein

1.30

berg

1.04

er

1.02

ers

1.01

hardt

1.01

enstein

0.92

bach

0.91

itz

0.86

bard

0.84

beck

0.80

Activations Density 0.015%

proper names or surnames related to a particular individual named Ben Stein

the name "Einstein" in various contexts

No Comments

No Known Activations