INDEX

Explanations

terms and phrases related to political and ethical discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÏİÎ½Î±

-0.18

ansi

-0.17

PEnd

-0.16

 PureComponent

-0.15

ynn

-0.15

oÅ¡

-0.14

.signals

-0.14

oÄį

-0.14

REW

-0.14

POSITIVE LOGITS

 accordingly

0.16

aight

0.15

anik

0.15

ivant

0.15

ubern

0.15

ORD

0.14

abei

0.14

Ord

0.14

nev

0.13

udge

0.13

Activations Density 0.167%

terms and phrases related to political and ethical discussions

No Comments

No Known Activations