INDEX

Explanations

social media posts or updates

oai_token-act-pair · gpt-3.5-turbo

references to photos and posts on social media

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anmar

-0.84

 Rounds

-0.77

senal

-0.73

osponsors

-0.71

metic

-0.67

 Centauri

-0.66

============

-0.65

fect

-0.63

enthal

-0.63

itone

-0.63

POSITIVE LOGITS

Ø

0.66

username

0.65

DragonMagazine

0.63

(@

0.62

0.61

atl

0.61

cov

0.61

0.59

 indu

0.59

ja

0.58

Activations Density 0.051%

social media posts or updates

references to photos and posts on social media

No Comments

No Known Activations