INDEX

Explanations

code-related terms and features of a user interface

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

lint

-0.19

SZ

-0.15

 Bullet

-0.14

 Legend

-0.14

 Industries

-0.14

ulnerable

-0.14

wner

-0.14

.ce

-0.14

otten

-0.13

à¸ļà¸²à¸¥

-0.13

POSITIVE LOGITS

éĴ®

0.17

errat

0.14

Ã©ny

0.14

_fa

0.14

.INSTANCE

0.14

)size

0.13

 Cann

0.13

atest

0.13

 molds

0.13

ava

0.13

Activations Density 0.022%

code-related terms and features of a user interface

No Comments

No Known Activations