INDEX

Explanations

technical terms and components related to programming and data structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Sizes

-0.26

 Values

-0.23

IDs

-0.21

 Ages

-0.21

Sizes

-0.20

 Names

-0.19

 Size

-0.17

Val

-0.17

 Rates

-0.17

 Titles

-0.17

POSITIVE LOGITS

 price

0.35

 name

0.28

age

0.25

(price

0.24

price

0.24

 weight

0.24

 title

0.23

(name

0.22

 value

0.22

 date

0.22

Activations Density 0.133%

technical terms and components related to programming and data structures

No Comments

No Known Activations