INDEX

Explanations

the name "Hag" and variations of it

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Hag" and related terms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

âķĲâķĲ

-0.71

Ô

-0.66

 Âł Âł Âł Âł

-0.64

 Samoa

-0.64

 Scotia

-0.63

âķĲ

-0.62

 Âł Âł Âł Âł Âł Âł Âł Âł

-0.62

 Advertisement

-0.62

ISION

-0.60

UTE

-0.59

POSITIVE LOGITS

glers

1.40

vest

0.91

nuts

0.91

rid

0.91

bard

0.88

raph

0.86

acity

0.85

elin

0.85

reen

0.85

masters

0.83

Activations Density 0.022%

the name "Hag" and variations of it

mentions of the name "Hag" and related terms

No Comments

No Known Activations