INDEX

Explanations

mentions of a specific name "Naruto"

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "Naruto"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ILCS

-0.99

mable

-0.76

 quarters

-0.73

boards

-0.73

strength

-0.73

drawn

-0.71

values

-0.71

blank

-0.71

med

-0.70

power

-0.70

POSITIVE LOGITS

uto

1.23

ÄŁ

1.00

Uz

0.96

zza

0.87

ctors

0.84

zzi

0.84

eca

0.78

rio

0.77

zin

0.75

á¹

0.75

Activations Density 0.021%

mentions of a specific name "Naruto"

mentions of the term "Naruto"

No Comments

No Known Activations