INDEX

Explanations

terms related to computational geometry and programming libraries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

eki

-0.15

-INF

-0.15

Mal

-0.14

.Av

-0.14

Ã«r

-0.14

ventional

-0.14

ucas

-0.14

JR

-0.14

 swear

-0.13

ëª©

-0.13

POSITIVE LOGITS

 BOOST

0.37

BOOST

0.33

 boost

0.33

 Boost

0.32

boost

0.30

Boost

0.28

<boost

0.27

.boost

0.26

(boost

0.25

 boosting

0.23

Activations Density 0.002%

terms related to computational geometry and programming libraries

No Comments

No Known Activations