INDEX

Explanations

references to specific foundations or organizations

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

.Writer

-0.16

angi

-0.16

fully

-0.15

ased

-0.15

phalt

-0.15

tered

-0.15

maxLength

-0.14

COP

-0.14

indent

-0.14

staking

-0.14

POSITIVE LOGITS

ry

0.32

ational

0.29

ries

0.26

ations

0.24

RY

0.23

amental

0.23

atio

0.22

lings

0.21

ling

0.21

rys

0.20

Activations Density 0.011%

references to specific foundations or organizations

No Comments

No Known Activations