INDEX

Explanations

tweets

oai_token-act-pair · gpt-3.5-turbo

instances of the word "tweeted" or related variations indicating social media activity

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

å£«

-0.74

cised

-0.68

bred

-0.65

 immersion

-0.62

å§«

-0.62

phal

-0.61

ricular

-0.61

 vein

-0.61

vantage

-0.59

 circumcision

-0.58

POSITIVE LOGITS

"@

0.81

weet

0.80

 hasht

0.80

realDonaldTrump

0.79

 condolences

0.78

yss

0.74

 angrily

0.72

storm

0.71

storms

0.70

 tweets

0.70

Activations Density 0.039%

tweets

instances of the word "tweeted" or related variations indicating social media activity

No Comments

No Known Activations