INDEX

Explanations

mentions of editorial boards and publications

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

Precision

-0.15

ti

-0.15

tant

-0.15

andid

-0.15

ammers

-0.14

elyn

-0.14

PR

-0.14

ÐµÐ¼Ð¾Ð½

-0.14

mony

-0.14

obi

-0.14

POSITIVE LOGITS

 newspaper

0.33

 Newspaper

0.26

 newspapers

0.22

ç´Ļ

0.22

 article

0.18

lesc

0.18

 crossword

0.17

 Newsp

0.17

 columnist

0.17

 mast

0.17

Activations Density 0.075%

mentions of editorial boards and publications

No Comments

No Known Activations