INDEX

Explanations

the name "Nancy Pelosi" mentioned in text

oai_token-act-pair · gpt-3.5-turbo

mentions of Nancy Pelosi

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

orescent

-0.78

adr

-0.74

tered

-0.74

igated

-0.72

oresc

-0.72

orical

-0.71

upon

-0.71

oning

-0.68

Spoiler

-0.66

inventoryQuantity

-0.66

POSITIVE LOGITS

 Pelosi

1.18

 Drew

0.96

 Grace

0.84

 McMahon

0.82

 Reagan

0.78

Lan

0.78

 Reed

0.75

Joy

0.75

 Loren

0.74

 Claus

0.74

Activations Density 0.008%

the name "Nancy Pelosi" mentioned in text

mentions of Nancy Pelosi

No Comments

No Known Activations