INDEX

Explanations

mentions of various individuals and their collaborative projects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

radu

-0.15

ãĥ¼ãĥŃ

-0.15

(BitConverter

-0.15

alar

-0.15

illet

-0.15

arga

-0.15

.toolbox

-0.15

Obr

-0.14

à¸ķà¸£à¸§

-0.14

okt

-0.14

POSITIVE LOGITS

Mas

0.27

 Hide

0.26

 Kenya

0.25

Nor

0.25

 Junk

0.24

 Minor

0.24

Hide

0.23

Kaz

0.23

Hi

0.23

Jun

0.23

Activations Density 0.035%

mentions of various individuals and their collaborative projects

No Comments

No Known Activations