INDEX

Explanations

direct speech or quotations within the text

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

 vault

-0.15

ovna

-0.14

iles

-0.14

-Smith

-0.14

uye

-0.14

.blogspot

-0.14

tsx

-0.13

meer

-0.13

à¤łà¤¨

-0.13

POSITIVE LOGITS

eyh

0.17

egin

0.14

zier

0.14

::-

0.14

ARIANT

0.14

 Instructor

0.14

 chilled

0.13

ynet

0.13

YST

0.13

 Trainer

0.13

Activations Density 0.062%

direct speech or quotations within the text

No Comments

No Known Activations