INDEX

Explanations

references to communication, specifically through calls, emails, and messages

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

 calls

-0.17

 calling

-0.16

_calls

-0.15

Ã¥n

-0.15

unm

-0.15

 Calls

-0.15

ombine

-0.15

 Against

-0.14

irk

-0.14

 remarks

-0.14

POSITIVE LOGITS

 letting

0.27

 asking

0.24

inform

0.23

 seeking

0.23

asking

0.22

 informing

0.22

 offering

0.21

 inviting

0.21

Inform

0.20

.request

0.20

Activations Density 0.166%

references to communication, specifically through calls, emails, and messages

No Comments

No Known Activations