INDEX

Explanations

instances of encryption-related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to cryptographic keys and security protocols

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

Fif

-0.65

 Colleges

-0.65

uum

-0.63

uate

-0.62

 simmer

-0.62

ews

-0.60

uated

-0.59

 Hanson

-0.59

asca

-0.59

ALLY

-0.59

POSITIVE LOGITS

stroke

1.49

chain

1.49

*/(

1.37

chains

1.31

bind

1.29

pad

1.24

caps

1.23

stro

1.23

stone

1.20

hole

1.19

Activations Density 0.033%

instances of encryption-related terms

terms related to cryptographic keys and security protocols

No Comments

No Known Activations