INDEX

Explanations

terms related to gender and gender identity issues

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

yan

-0.16

vip

-0.15

_RING

-0.14

sher

-0.14

sit

-0.14

oted

-0.14

deo

-0.14

arness

-0.14

ycz

-0.14

lik

-0.13

POSITIVE LOGITS

ed

0.40

edn

0.26

que

0.26

-neutral

0.25

 roles

0.25

 neutral

0.24

neutral

0.24

fluid

0.23

ED

0.22

 equality

0.22

Activations Density 0.012%

terms related to gender and gender identity issues

No Comments

No Known Activations