INDEX

Explanations

locations or areas considered as strongholds for various groups or entities

oai_token-act-pair · gpt-3.5-turbo

references to strategic locations or bases of influence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

nih

-0.79

erson

-0.76

respond

-0.73

istry

-0.72

hy

-0.72

umn

-0.71

andro

-0.70

abus

-0.70

endment

-0.69

 Blossom

-0.68

POSITIVE LOGITS

 stronghold

1.01

 strongh

0.95

holder

0.82

hold

0.80

urst

0.75

fast

0.72

 ende

0.71

 indo

0.71

 tradem

0.70

 concentrated

0.69

Activations Density 0.025%

locations or areas considered as strongholds for various groups or entities

references to strategic locations or bases of influence

No Comments

No Known Activations