INDEX

Explanations

terms associated with LGBTQ+ identities and pride events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

ÑĢÑĥÑĪ

-0.18

ackages

-0.15

 swingers

-0.15

amp

-0.14

kan

-0.14

.Mask

-0.14

aden

-0.14

 Sexo

-0.13

áº¿n

-0.13

bro

-0.13

POSITIVE LOGITS

/trans

0.18

-friendly

0.18

QRSTUV

0.16

ãĥ¼ãĥ³

0.16

 spectrum

0.16

 pride

0.15

 friendly

0.15

tik

0.15

 rights

0.15

QQ

0.15

Activations Density 0.035%

terms associated with LGBTQ+ identities and pride events

No Comments

No Known Activations

terms associated with LGBTQ+ identities and pride events

No Comments

No Known Activations