INDEX

Explanations

expressions of appreciation or desire

oai_token-act-pair · gpt-3.5-turbo

expressions of desire or preference

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

VERTISEMENT

-0.86

advertisement

-0.68

è¦ļéĨĴ

-0.67

 psyche

-0.63

rift

-0.62

 Accessed

-0.62

partisan

-0.61

 divides

-0.60

half

-0.60

 consumes

-0.57

POSITIVE LOGITS

lihood

0.75

 Asylum

0.73

lier

0.72

yip

0.70

istani

0.69

liest

0.68

 DACA

0.61

iewicz

0.61

erto

0.60

xus

0.60

Activations Density 0.062%

expressions of appreciation or desire

expressions of desire or preference

No Comments

No Known Activations