INDEX

Explanations

terms related to counting and measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

kind

-0.18

ated

-0.17

curity

-0.17

ettings

-0.16

olis

-0.16

 ç©

-0.15

 váºŃy

-0.15

ccount

-0.15

ity

-0.15

/current

-0.15

POSITIVE LOGITS

enance

0.42

DownLatch

0.30

inue

0.26

erc

0.24

ers

0.23

ering

0.23

ains

0.22

ants

0.22

erv

0.22

INUE

0.22

Activations Density 0.028%

terms related to counting and measurements

No Comments

No Known Activations