INDEX

Explanations

names of people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ecom

-0.17

CHIP

-0.15

optic

-0.15

Å¾it

-0.15

Ã¡tu

-0.14

adla

-0.14

è¿

-0.14

KHTML

-0.14

qml

-0.13

meric

-0.13

POSITIVE LOGITS

-tm

0.15

 hypo

0.15

-carousel

0.14

 Ryder

0.14

 spread

0.14

 Motor

0.13

Lah

0.13

cop

0.13

ania

0.13

rott

0.13

Activations Density 0.254%

names of people and places

No Comments

No Known Activations