INDEX

Explanations

negative phrases or contractions indicating 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 @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

okt

-0.16

Äįku

-0.15

infeld

-0.15

.Framework

-0.15

uler

-0.15

lider

-0.14

QueryParam

-0.14

WebHost

-0.14

raphics

-0.14

ioxide

-0.14

POSITIVE LOGITS

 worry

0.24

let

0.23

cha

0.23

 forget

0.23

cha

0.22

 despair

0.20

 expect

0.18

 fret

0.18

CHA

0.17

 ever

0.17

Activations Density 0.042%

negative phrases or contractions indicating refusal or denial

No Comments

No Known Activations