INDEX

Explanations

references to queer identity and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

nya

-0.18

shape

-0.18

ship

-0.16

orne

-0.16

son

-0.16

sw

-0.16

ly

-0.16

sun

-0.16

-0.15

li

-0.15

POSITIVE LOGITS

uing

0.29

ued

0.28

bec

0.28

ues

0.28

ens

0.21

erness

0.21

uetype

0.20

UES

0.19

estion

0.19

ENS

0.17

Activations Density 0.008%

references to queer identity and related terms

No Comments

No Known Activations