INDEX

Explanations

actions that imply physical interaction or violence

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

jÃŃt

-0.15

itten

-0.15

ijd

-0.14

abela

-0.14

 mind

-0.14

izzo

-0.13

vfs

-0.13

ILLA

-0.13

iom

-0.13

okino

-0.13

POSITIVE LOGITS

 Earn

0.17

lege

0.15

æĴĥ

0.14

 clave

0.14

ients

0.14

ekim

0.14

awe

0.14

udas

0.14

zahl

0.14

Activations Density 0.187%

actions that imply physical interaction or violence

No Comments

No Known Activations