INDEX

Explanations

phrases that indicate refusal or denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

eni

-0.16

 Ø³ÙĦØ§Ùħ

-0.16

ewe

-0.16

abus

-0.15

ekim

-0.15

onda

-0.15

erdem

-0.15

afe

-0.15

patch

-0.14

etto

-0.14

POSITIVE LOGITS

let

0.20

 admit

0.18

 accept

0.18

 allow

0.17

 accepting

0.17

 letting

0.16

 Accept

0.16

Let

0.16

Accept

0.16

 accepts

0.15

Activations Density 0.117%

phrases that indicate refusal or denial

No Comments

No Known Activations