INDEX

Explanations

instances of personal names and connections to specific events or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Higgins

-0.17

Fi

-0.16

zak

-0.15

Ø§ÙĪÙĩ

-0.14

 Hacker

-0.14

ãĤ¶ãĥ¼

-0.14

Corp

-0.14

ugal

-0.14

uno

-0.13

liga

-0.13

POSITIVE LOGITS

 Arthropoda

0.16

Www

0.15

Mos

0.15

AMAGE

0.15

otec

0.15

Sov

0.15

Nev

0.15

âħ

0.14

erland

0.14

 Beitrag

0.14

Activations Density 0.304%

instances of personal names and connections to specific events or roles

No Comments

No Known Activations