INDEX

Explanations

locations around the world

oai_token-act-pair · gpt-3.5-turbo

phrases indicating possession or attributes related to various subjects or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

alter

-0.73

FG

-0.63

 IMAGES

-0.63

matter

-0.63

iversary

-0.62

oyer

-0.61

 Explain

-0.61

arters

-0.60

ohm

-0.60

TAG

-0.59

POSITIVE LOGITS

 undergone

1.21

 become

1.18

 witnessed

1.09

 been

1.08

 seen

1.01

 suffered

0.98

 endured

0.96

 struggled

0.94

 fared

0.94

 benefited

0.92

Activations Density 0.160%

locations around the world

phrases indicating possession or attributes related to various subjects or locations

No Comments

No Known Activations