INDEX

Explanations

various concepts and ideas related to understanding and interpretation

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

uros

-0.15

 ÙĪØ£ÙĨ

-0.15

illac

-0.14

odule

-0.14

ynet

-0.14

ture

-0.14

ivec

-0.14

hiba

-0.13

BN

-0.13

loth

-0.13

POSITIVE LOGITS

how

0.41

 what

0.32

as

0.31

how

0.29

why

0.28

 about

0.28

 cÃ³mo

0.26

HOW

0.25

what

0.24

on

0.22

Activations Density 0.137%

various concepts and ideas related to understanding and interpretation

No Comments

No Known Activations