INDEX

Explanations

words related to investigation or enhancement

oai_token-act-pair · gpt-3.5-turbo

words related to intense emotional states or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Kush

-0.67

 aisle

-0.64

 uneven

-0.63

liners

-0.60

 Cinema

-0.60

 driveway

-0.58

 runway

-0.58

 spotting

-0.58

air

-0.58

 Circus

-0.56

POSITIVE LOGITS

ment

0.90

¶æ

0.88

orial

0.81

vous

0.80

ments

0.78

enced

0.78

uel

0.78

orate

0.77

ement

0.77

abil

0.77

Activations Density 0.191%

words related to investigation or enhancement

words related to intense emotional states or experiences

No Comments

No Known Activations