INDEX

Explanations

phrases related to rewards or compensation

oai_token-act-pair · gpt-3.5-turbo

phrases related to rewards or compensation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

soDeliveryDate

-1.00

edin

-0.76

ãģ®éŃĶ

-0.74

DragonMagazine

-0.72

stanbul

-0.70

ãĤ¼ãĤ¦ãĤ¹

-0.69

neapolis

-0.67

ise

-0.67

ulture

-0.66

legraph

-0.65

POSITIVE LOGITS

 violating

1.08

 sins

1.01

 completing

1.00

 complying

0.99

 failing

0.99

bidden

0.97

 defeating

0.97

 inaction

0.96

gery

0.95

 breaching

0.94

Activations Density 0.193%

phrases related to rewards or compensation

phrases related to rewards or compensation

No Comments

No Known Activations