INDEX

Explanations

references to tombs

oai_token-act-pair · gpt-3.5-turbo

references to tombs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

-0.93

Ã±a

-0.79

Equ

-0.79

ÃĽ

-0.77

IGHTS

-0.74

CRE

-0.74

Population

-0.73

yip

-0.73

CE

-0.71

OHN

-0.71

POSITIVE LOGITS

stones

1.26

stone

1.19

 tomb

1.10

 Raider

0.98

 Tomb

0.90

 grave

0.84

 vault

0.82

 robbers

0.81

tom

0.81

ologic

0.80

Activations Density 0.007%

references to tombs

references to tombs

No Comments

No Known Activations