INDEX

Explanations

references to Washington or its institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ymax

-0.15

yw

-0.15

ertation

-0.14

ãĥ©ãĥĥãĤ¯

-0.14

ei

-0.14

Ø§ÙĨÙĩ

-0.14

Ð´Ð°Ð¼

-0.14

ãĥ¼ãĤ

-0.14

ilk

-0.14

owell

-0.14

POSITIVE LOGITS

DC

0.31

0.26

DC

0.25

 Redskins

0.23

dc

0.23

dc

0.22

 Irving

0.22

inton

0.21

(dc

0.20

 Wash

0.19

Activations Density 0.022%

references to Washington or its institutions

No Comments

No Known Activations