INDEX

Explanations

phrases related to leadership roles and organizational positions

oai_token-act-pair · gpt-3.5-turbo

references to individuals in positions of authority or leadership

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

tains

-0.71

 Flavoring

-0.67

fits

-0.64

Vers

-0.63

Runner

-0.63

Posts

-0.60

 certs

-0.60

 Lives

-0.60

CLASS

-0.60

Translation

-0.58

POSITIVE LOGITS

 intend

1.14

 acknowledge

1.11

 concede

1.10

 insist

1.10

 contend

1.04

 allege

1.01

are

1.00

say

0.99

 have

0.99

 admit

0.97

Activations Density 0.210%

phrases related to leadership roles and organizational positions

references to individuals in positions of authority or leadership

No Comments

No Known Activations