INDEX

Explanations

references to dates and publication details

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

agnost

-0.18

Ð²Ð°Ð¹

-0.15

luet

-0.15

blers

-0.15

 Lomb

-0.15

itur

-0.15

 íĥĪ

-0.14

 Woodward

-0.14

Î¯Î½Î·

-0.14

adeon

-0.14

POSITIVE LOGITS

dech

0.17

ushman

0.15

hoot

0.15

.Ret

0.14

mouseout

0.14

asted

0.13

 pictured

0.13

Ð¾Ð·

0.13

âĢİ

0.13

 northeastern

0.13

Activations Density 0.405%

references to dates and publication details

No Comments

No Known Activations