INDEX

Explanations

phrases related to personal or professional connections, specifically focusing on the term 'associates'

oai_token-act-pair · gpt-3.5-turbo

mentions of "associates" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Ott

-0.71

 Emperor

-0.70

 Levant

-0.70

 Patriarch

-0.68

Politics

-0.66

Fly

-0.66

=-=-=-=-=-=-=-=-

-0.66

bra

-0.65

blank

-0.65

Louis

-0.64

POSITIVE LOGITS

 associates

1.38

 associate

1.06

ociate

0.90

iated

0.81

 associations

0.80

iate

0.79

leground

0.77

hips

0.75

 warr

0.74

 affiliates

0.74

Activations Density 0.005%

phrases related to personal or professional connections, specifically focusing on the term 'associates'

mentions of "associates" in various contexts

No Comments

No Known Activations