INDEX

Explanations

references to technological processes or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

æĬľ

-0.18

udoku

-0.15

ogl

-0.14

Lad

-0.14

ãĥ«ãĥī

-0.14

wort

-0.14

 Shank

-0.14

SPI

-0.14

 Ð·Ð°Ð¹

-0.14

»¿

-0.14

POSITIVE LOGITS

 motor

0.23

 tela

0.21

mot

0.21

motor

0.20

dot

0.20

 Motor

0.19

co

0.19

Motor

0.18

 mont

0.17

 mount

0.17

Activations Density 0.005%

references to technological processes or systems

No Comments

No Known Activations