INDEX

Explanations

references to individuals such as Podesta, Hillary Clinton, and Huma Abedin

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals associated with a political context, particularly focusing on the name "Podesta."

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

liness

-0.76

lights

-0.75

 Jericho

-0.75

lasses

-0.74

lessly

-0.73

TRY

-0.73

breaks

-0.69

phal

-0.66

mental

-0.65

iveness

-0.63

POSITIVE LOGITS

 Podesta

1.04

esta

0.90

 Emails

0.87

email

0.79

artisan

0.77

 Address

0.76

agate

0.74

fman

0.74

afort

0.72

ovo

0.71

Activations Density 0.016%

references to individuals such as Podesta, Hillary Clinton, and Huma Abedin

mentions of specific individuals associated with a political context, particularly focusing on the name "Podesta."

No Comments

No Known Activations