INDEX

Explanations

references to picking or selections

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

inea

-0.16

iously

-0.16

ibble

-0.15

 pity

-0.15

 Keys

-0.15

 apart

-0.15

amework

-0.15

ãĥ©ãĤ¹

-0.14

á»ĩu

-0.14

ãĥ¼ãĥĸ

-0.14

POSITIVE LOGITS

led

0.29

les

0.28

ering

0.28

wick

0.25

axe

0.24

achu

0.23

axes

0.23

ERING

0.22

lem

0.22

lesh

0.21

Activations Density 0.007%

references to picking or selections

No Comments

No Known Activations