INDEX

Explanations

HTML and structured content elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

lyph

-0.14

LabelText

-0.14

 Hood

-0.13

abar

-0.13

Moo

-0.13

Cle

-0.13

Muj

-0.13

Ã¬m

-0.13

eci

-0.13

Ø§ÙĪÙĨØ¯

-0.13

POSITIVE LOGITS

.hr

0.22

ulist

0.22

hr

0.21

HR

0.20

HR

0.17

hr

0.17

UL

0.17

UL

0.17

.quote

0.16

<p

0.16

Activations Density 0.068%

HTML and structured content elements

No Comments

No Known Activations