INDEX

Explanations

references to having meals or dining with others

oai_token-act-pair · gpt-3.5-turbo

phrases indicating social interactions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mart

-0.78

©¶æ¥µ

-0.73

mite

-0.71

ghai

-0.68

fighter

-0.68

arily

-0.66

ovich

-0.63

nery

-0.63

æ©Ł

-0.62

sov

-0.62

POSITIVE LOGITS

 regards

1.17

stood

1.08

 impunity

1.08

 regard

1.04

standing

1.01

draw

0.99

 dolls

0.89

 dignity

0.85

 dolphins

0.82

holding

0.82

Activations Density 0.202%

references to having meals or dining with others

phrases indicating social interactions

No Comments

No Known Activations