INDEX

Explanations

phrases related to individuals named "Jo" followed by a number

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

flies

-0.68

 interests

-0.67

IFIED

-0.67

 Fired

-0.65

BALL

-0.64

raits

-0.64

 IMAGES

-0.62

Pwr

-0.62

 amplification

-0.62

CLASS

-0.62

POSITIVE LOGITS

aquin

1.40

Jo

1.23

ining

1.08

zeb

1.07

anne

1.07

jo

1.06

zy

1.01

ppa

1.01

areth

0.97

Anne

0.96

Activations Density 0.018%

phrases related to individuals named "Jo" followed by a number

No Comments

No Known Activations