INDEX

Explanations

phrases related to refusals and offers of assistance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

oste

-0.16

 succesfully

-0.16

abilidad

-0.15

avia

-0.15

||(

-0.15

omaly

-0.15

erca

-0.15

ãĤ¸ãĤ¢

-0.15

 lesbienne

-0.14

lexport

-0.14

POSITIVE LOGITS

let

0.30

 accept

0.29

 accepting

0.28

 allow

0.27

 letting

0.26

 accepts

0.26

 Accept

0.25

accept

0.24

 allowing

0.23

let

0.22

Activations Density 0.136%

phrases related to refusals and offers of assistance

No Comments

No Known Activations