INDEX

Explanations

occurrences of the name "Kurt"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Kurt" and references to the band Nirvana

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

Capital

-0.75

cale

-0.74

advertisement

-0.70

dress

-0.66

cyn

-0.66

 circ

-0.66

ersion

-0.65

wrapper

-0.64

ebus

-0.64

eches

-0.63

POSITIVE LOGITS

 Kurt

1.26

Von

1.21

zman

1.09

 Angle

0.92

Cob

0.88

 Hein

0.87

 Schwarz

0.86

ãĥĦ

0.84

 Braun

0.80

 Ares

0.78

Activations Density 0.003%

occurrences of the name "Kurt"

mentions of the name "Kurt" and references to the band Nirvana

No Comments

No Known Activations