INDEX

Explanations

references to subatomic particles and their properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ris

-0.17

iar

-0.16

butt

-0.15

_endian

-0.15

licht

-0.15

abo

-0.14

si

-0.14

ACES

-0.14

æķ

-0.14

alian

-0.14

POSITIVE LOGITS

 neutr

0.28

 charged

0.25

 neut

0.24

 charm

0.23

qu

0.21

 electrons

0.21

rons

0.21

charm

0.21

uds

0.20

 Charm

0.20

Activations Density 0.028%

references to subatomic particles and their properties

No Comments

No Known Activations