INDEX

Explanations

terms related to social media recommendation algorithms

oai_token-act-pair · gpt-3.5-turbo

references to social media interactions and community engagements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

amination

-0.69

constitutional

-0.69

MpServer

-0.68

Cth

-0.67

uld

-0.67

 miscon

-0.64

byte

-0.64

NEC

-0.62

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.62

atism

-0.62

POSITIVE LOGITS

 curated

1.18

 notifications

1.18

 personalized

1.10

 favorites

1.06

 Favor

1.00

 Moments

1.00

 chats

0.98

 emoji

0.95

 hasht

0.95

 trending

0.93

Activations Density 0.532%

terms related to social media recommendation algorithms

references to social media interactions and community engagements

No Comments

No Known Activations