INDEX

Explanations

references to famous musical bands and artists such as Nirvana and The Beatles

oai_token-act-pair · gpt-3.5-turbo

mentions of iconic music bands, particularly Nirvana and The Beatles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

uler

-0.78

bot

-0.77

eus

-0.75

sab

-0.69

ebus

-0.67

rongh

-0.66

bay

-0.65

izon

-0.64

ouri

-0.63

uron

-0.61

POSITIVE LOGITS

 Beatles

1.91

 Nirvana

1.90

 Billboard

1.73

 McCartney

1.71

 Lennon

1.71

 frontman

1.69

 Metall

1.64

 guitarist

1.61

 Eminem

1.59

 Madonna

1.55

Activations Density 0.113%

references to famous musical bands and artists such as Nirvana and The Beatles

mentions of iconic music bands, particularly Nirvana and The Beatles

No Comments

No Known Activations