INDEX

Explanations

references to acts of covering or concealment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

acl

-0.15

byt

-0.14

createFrom

-0.14

è¶ħ

-0.14

ad

-0.14

 Everyday

-0.14

 heels

-0.14

POSITIVE LOGITS

 ÑģÐ¾Ð±Ð¾Ñİ

0.23

 ÑģÐ¾Ð±Ð¾Ð¹

0.22

 entire

0.19

 naked

0.18

 Entire

0.17

Ð¸Ð½ÑĥÐ²

0.16

 mouth

0.15

 areas

0.15

 Naked

0.15

 bare

0.15

Activations Density 0.111%

references to acts of covering or concealment

No Comments

No Known Activations