INDEX

Explanations

words or phrases related to listing or detailing various items or aspects

oai_token-act-pair · gpt-3.5-turbo

references to general topics or issues being discussed

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bern

-0.79

 Riders

-0.66

inav

-0.64

RU

-0.62

igun

-0.62

 rematch

-0.61

onz

-0.60

istar

-0.59

VG

-0.59

gaard

-0.59

POSITIVE LOGITS

 transpired

0.79

worldly

0.76

nces

0.73

hots

0.71

 includ

0.70

 challeng

0.68

lihood

0.68

zyme

0.67

 pertaining

0.65

hooting

0.65

Activations Density 0.016%

words or phrases related to listing or detailing various items or aspects

references to general topics or issues being discussed

No Comments

No Known Activations