INDEX

Explanations

references to scientific terms and concepts related to research and classification

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

leyen

-0.17

 Erotische

-0.16

zens

-0.16

RGBA

-0.16

innacle

-0.15

 Welch

-0.15

gaard

-0.15

 Stap

-0.15

adero

-0.15

sey

-0.14

POSITIVE LOGITS

iah

0.18

Kar

0.17

olu

0.17

ht

0.17

 isol

0.16

 Tune

0.16

imer

0.16

Rad

0.16

æķ¦

0.16

cou

0.15

Activations Density 0.005%

references to scientific terms and concepts related to research and classification

No Comments

No Known Activations