INDEX

Explanations

any word related to the Roman figure Antoninus

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

glers

-0.97

Kit

-0.71

lder

-0.67

FU

-0.67

ãĤĭ

-0.64

Privacy

-0.64

gery

-0.63

DIV

-0.63

tc

-0.62

eled

-0.62

POSITIVE LOGITS

opoulos

1.10

inus

1.02

aic

1.01

ique

1.00

inian

0.99

opol

0.98

iol

0.97

iques

0.97

ion

0.95

 Anton

0.94

Activations Density 0.017%

any word related to the Roman figure Antoninus

No Comments

No Known Activations