INDEX

Explanations

phrases related to legal or criminal activities and regulatory terms

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

ccording

-0.85

ALK

-0.81

agonists

-0.75

enhagen

-0.75

NEW

-0.75

VL

-0.75

PDATE

-0.74

aution

-0.74

vertisement

-0.74

GoldMagikarp

-0.73

POSITIVE LOGITS

qui

1.20

 Ãł

1.14

que

1.07

du

1.02

 pour

1.01

des

0.96

aux

0.96

est

0.96

et

0.95

 mund

0.92

Activations Density 0.111%

phrases related to legal or criminal activities and regulatory terms

No Comments

No Known Activations