INDEX

Explanations

questions starting with "How"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

.timeScale

-0.14

æ¹¯

-0.14

âĢİ

-0.14

/fontawesome

-0.14

 ë¶Ģíĥģ

-0.13

onse

-0.13

inyin

-0.13

 htmlFor

-0.13

/null

-0.13

.heroku

-0.13

POSITIVE LOGITS

do

0.35

 does

0.35

 Does

0.32

Do

0.32

did

0.29

To

0.28

 Much

0.28

rah

0.27

dy

0.27

can

0.26

Activations Density 0.041%

questions starting with "How"

No Comments

No Known Activations