INDEX

Explanations

phrases indicating intent or significance

oai_token-act-pair · gpt-3.5-turbo

phrases that express the significance or value of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ioxide

-0.71

Myth

-0.69

pse

-0.62

sbm

-0.62

amide

-0.60

hari

-0.60

 Myth

-0.60

 manufact

-0.59

 Notting

-0.58

 faculties

-0.58

POSITIVE LOGITS

uni

0.78

terday

0.76

ranged

0.74

plain

0.72

ortium

0.69

thood

0.69

Ã±a

0.68

 disrespect

0.68

Ã±

0.65

 farewell

0.65

Activations Density 0.086%

phrases indicating intent or significance

phrases that express the significance or value of something

No Comments

No Known Activations