INDEX

Explanations

references to social media posts, particularly tweets

oai_token-act-pair · gpt-3.5-turbo

instances of the word "tweet."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sbm

-0.73

ajor

-0.69

hibited

-0.68

ccording

-0.67

cum

-0.66

vantage

-0.65

NetMessage

-0.65

pure

-0.64

 Penet

-0.63

 Brill

-0.63

POSITIVE LOGITS

 tweet

1.08

storms

1.07

 Tweet

1.05

storm

1.00

Tweet

1.00

 hashtag

0.94

 tweets

0.93

deck

0.89

aires

0.80

 emoji

0.80

Activations Density 0.007%

references to social media posts, particularly tweets

instances of the word "tweet."

No Comments

No Known Activations