INDEX

Explanations

phrases related to safety or alert systems

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Turn" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

è¦ļéĨĴ

-0.76

ãģª

-0.74

ãĤĮ

-0.69

ãģ®ç

-0.64

ASC

-0.64

 Lama

-0.63

 shorth

-0.63

 Kahn

-0.63

ãģŁ

-0.62

atorial

-0.62

POSITIVE LOGITS

turn

1.09

Turn

1.00

 Turn

0.83

bum

0.78

overs

0.77

ners

0.75

wheel

0.74

 Turns

0.73

 inward

0.71

ueless

0.71

Activations Density 0.007%

phrases related to safety or alert systems

instances of the word "Turn" in various contexts

No Comments

No Known Activations