INDEX

Explanations

names of individuals or significant group titles related to responsibility or leadership

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ãĢģ

-0.26

ãĢģä¸Ń

-0.25

etc

-0.24

-,

-0.23

ãĢģå¤§

-0.23

etc

-0.23

 ãĢģ

-0.21

ãĢģãĢĬ

-0.20

ãĢģé«ĺ

-0.20

ãĢģå°ı

-0.20

POSITIVE LOGITS

AndPassword

0.16

ä¸İ

0.15

 resp

0.15

èĪĩ

0.15

",-

0.14

ÑĢÐ¾Ñĩ

0.14

.scalablytyped

0.14

ëĬĲ

0.14

_exempt

0.14

YE

0.14

Activations Density 0.268%

names of individuals or significant group titles related to responsibility or leadership

No Comments

No Known Activations