INDEX

Explanations

references to historical performances or reenactments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

 Drain

-0.15

 Boots

-0.15

íĥ

-0.15

quer

-0.14

ëĶ

-0.14

ÑģÐ°Ð¼

-0.14

adb

-0.14

 ÑģÐºÐ°

-0.14

ä»ĭ

-0.13

agraph

-0.13

POSITIVE LOGITS

illard

0.18

oust

0.16

asca

0.16

laws

0.15

alerts

0.15

å¾Ģ

0.15

0.14

unate

0.14

erties

0.14

iest

0.14

Activations Density 0.150%

references to historical performances or reenactments

No Comments

No Known Activations