INDEX

Explanations

words related to programming, software, and technical specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

aux

-0.23

Ð²Ð°Ð½Ð½Ñı

-0.20

à¸£à¸²à¸¢

-0.19

istic

-0.19

à¸£

-0.18

à¸£à¸§à¸¡

-0.17

à¸¥à¸¥

-0.17

ÌĨ

-0.17

esti

-0.17

ê·¹

-0.16

POSITIVE LOGITS

tober

0.35

nowledge

0.32

nowled

0.30

kk

0.26

owski

0.25

à¹Ģà¸ģà¸Ńà¸£

0.25

ansas

0.25

kel

0.24

intosh

0.23

ety

0.23

Activations Density 0.671%

words related to programming, software, and technical specifications

No Comments

No Known Activations