INDEX

Explanations

references to specific actions or roles performed by individuals

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soDeliveryDate

-0.78

 Groups

-0.68

wang

-0.66

iland

-0.64

Menu

-0.62

paces

-0.62

 Nicol

-0.62

 menus

-0.61

alde

-0.59

hatt

-0.58

POSITIVE LOGITS

 sins

1.31

 inaction

1.15

 transgress

1.06

 crimes

1.05

 mistakes

1.04

 actions

1.01

 sake

1.00

 failures

0.99

 neglect

0.97

 negligence

0.97

Activations Density 0.404%

references to specific actions or roles performed by individuals

No Comments

No Known Activations