INDEX

Explanations

phrases related to software functionalities and capabilities

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the ability or permissions granted to the user

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

falls

-0.64

stood

-0.57

 Winning

-0.56

 Worth

-0.55

ukemia

-0.52

aval

-0.52

igma

-0.51

 Supporters

-0.51

xon

-0.51

outed

-0.50

POSITIVE LOGITS

to

1.06

 customize

1.00

 specify

0.91

to

0.85

 configure

0.83

 automate

0.83

 manipulate

0.78

 choose

0.78

 visualize

0.77

 browse

0.77

Activations Density 0.124%

phrases related to software functionalities and capabilities

phrases indicating the ability or permissions granted to the user

No Comments

No Known Activations