INDEX

Explanations

references to authorship and contributors in artistic or academic contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

anske

-0.17

 LOCK

-0.16

oyer

-0.16

utin

-0.15

 ÎĬ

-0.15

 Scho

-0.15

 uplift

-0.15

LETE

-0.15

Ø¸Ùģ

-0.15

-fw

-0.15

POSITIVE LOGITS

ilib

0.15

icias

0.14

ãĥ¼ãĥ

0.14

 Angel

0.14

ddy

0.14

dd

0.13

/*č↵

0.13

rahim

0.13

ffer

0.13

Angel

0.13

Activations Density 0.150%

references to authorship and contributors in artistic or academic contexts

No Comments

No Known Activations