INDEX

Explanations

references to sensitive societal issues, particularly relating to consent and interpersonal relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

uide

-0.16

amma

-0.15

ius

-0.15

aleb

-0.14

sch

-0.14

 your

-0.14

jour

-0.14

akk

-0.14

 touch

-0.14

 inters

-0.14

POSITIVE LOGITS

 sometimes

0.27

sometimes

0.26

usually

0.25

variably

0.24

 usually

0.24

ometimes

0.24

often

0.22

 often

0.21

 either

0.20

 Ð»Ð¸Ð±Ð¾

0.20

Activations Density 0.308%

references to sensitive societal issues, particularly relating to consent and interpersonal relationships

No Comments

No Known Activations