INDEX

Explanations

phrases indicating willingness or refusal to take actions

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

avia

-0.17

ÐľÐŀ

-0.16

åĨ³å®ļ

-0.16

 rÃ©uss

-0.14

ysi

-0.14

 beforeSend

-0.14

_ACL

-0.14

 succesfully

-0.14

.ribbon

-0.14

emey

-0.14

POSITIVE LOGITS

 accept

0.25

 accepting

0.24

 accepts

0.22

let

0.22

 Accept

0.21

 compromise

0.21

 admit

0.20

accept

0.20

 cooperation

0.19

 cooperate

0.19

Activations Density 0.100%

phrases indicating willingness or refusal to take actions

No Comments

No Known Activations