INDEX

Explanations

terms related to different parts of the body

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Vide

-0.77

 Berks

-0.68

UE

-0.63

 Argon

-0.63

 Restaur

-0.63

NESS

-0.63

 Forbidden

-0.62

actionDate

-0.62

Soy

-0.62

Suc

-0.62

POSITIVE LOGITS

wash

1.16

bones

1.14

piece

1.08

guards

1.08

bone

1.07

ache

1.03

 circumference

1.00

shake

0.99

plant

0.99

washing

0.99

Activations Density 3.772%

terms related to different parts of the body

No Comments

No Known Activations