INDEX

Explanations

references to biblical or spiritual imagery

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

umo

-0.15

riba

-0.14

rometer

-0.14

oksen

-0.14

unken

-0.14

marvin

-0.14

"crypto

-0.14

nul

-0.13

enu

-0.13

POSITIVE LOGITS

0.18

qa

0.16

vir

0.14

 kaliteli

0.14

 pestic

0.14

 Resident

0.14

æ¡Ĥ

0.14

 task

0.14

 Plex

0.13

acha

0.13

Activations Density 0.312%

references to biblical or spiritual imagery

No Comments

No Known Activations