INDEX

Explanations

words related to authenticity or authentication

oai_token-act-pair · gpt-3.5-turbo

references to authenticity and genuine identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Pharma

-0.67

armac

-0.63

Vert

-0.63

Pages

-0.62

storms

-0.61

Population

-0.61

Mp

-0.59

pai

-0.59

olyn

-0.59

acent

-0.58

POSITIVE LOGITS

 authentic

0.99

icator

0.94

 authenticity

0.91

authent

0.88

 authent

0.81

uine

0.80

 Authent

0.80

ibly

0.80

imated

0.80

ãĥīãĥ©ãĤ´ãĥ³

0.76

Activations Density 0.022%

words related to authenticity or authentication

references to authenticity and genuine identity

No Comments

No Known Activations