INDEX

Explanations

capitalized words

oai_token-act-pair · gpt-3.5-turbo

specific letter combinations or patterns within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Magikarp

-0.74

yip

-0.66

 Clover

-0.63

clinton

-0.62

DragonMagazine

-0.61

chev

-0.59

 Twain

-0.58

Ô

-0.57

channelAvailability

-0.57

Bog

-0.56

POSITIVE LOGITS

asion

0.88

inct

0.83

ributed

0.81

resso

0.79

achment

0.78

idential

0.77

ached

0.77

ossession

0.77

erves

0.75

itt

0.74

Activations Density 0.110%

capitalized words

specific letter combinations or patterns within text

No Comments

No Known Activations