INDEX

Explanations

references to dairy products

oai_token-act-pair · gpt-3.5-turbo

references to dairy products

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

inx

-0.74

undo

-0.73

itud

-0.69

itudinal

-0.68

 Antar

-0.68

 comprehension

-0.66

 Colt

-0.65

htar

-0.65

ĺħ

-0.65

arters

-0.64

POSITIVE LOGITS

ious

0.93

lig

0.79

ried

0.76

ries

0.75

inity

0.74

minist

0.71

 Goods

0.70

nation

0.70

 herds

0.69

rier

0.69

Activations Density 0.049%

references to dairy products

references to dairy products

No Known Activations