INDEX

Explanations

terminology and concepts related to network analysis and community structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vik

-0.16

 Bracket

-0.15

rell

-0.15

iyat

-0.14

ãĥªãĤ¹

-0.14

 Loft

-0.14

 Monument

-0.14

ÎºÏħ

-0.14

logan

-0.14

 bure

-0.14

POSITIVE LOGITS

 network

0.30

 networks

0.28

 node

0.27

 Watts

0.27

network

0.26

 assort

0.26

 Network

0.25

Erd

0.24

 nodes

0.24

Network

0.24

Activations Density 0.228%

terminology and concepts related to network analysis and community structures

No Comments

No Known Activations