INDEX

Explanations

phrases containing the word 'hero'

oai_token-act-pair · gpt-3.5-turbo

the term "ero" which is likely related to defense or an industry sector

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ividual

-0.75

hips

-0.73

lv

-0.69

eenth

-0.66

 Siber

-0.65

liam

-0.65

ivities

-0.64

grave

-0.63

igators

-0.63

ership

-0.62

POSITIVE LOGITS

edia

1.01

ero

0.86

eli

0.86

oma

0.82

usalem

0.81

ose

0.79

oya

0.79

bably

0.79

zzi

0.79

cephal

0.78

Activations Density 0.009%

phrases containing the word 'hero'

the term "ero" which is likely related to defense or an industry sector

No Comments

No Known Activations