INDEX

Explanations

phrases where the speaker or someone else is mentioned

oai_token-act-pair · gpt-3.5-turbo

references to the speaker or personal experiences

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

*/(

-0.73

roup

-0.66

itect

-0.65

trump

-0.65

ulton

-0.64

leton

-0.63

JUST

-0.63

osterone

-0.63

inelli

-0.63

iston

-0.63

POSITIVE LOGITS

adow

1.00

adows

0.99

zzo

0.99

lees

0.88

imei

0.82

 suppose

0.81

opia

0.80

agi

0.77

 personally

0.75

 territ

0.71

Activations Density 0.071%

phrases where the speaker or someone else is mentioned

references to the speaker or personal experiences

No Comments

No Known Activations