INDEX

Explanations

methods and strategies related to best practices in various contexts

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

heid

-0.16

vem

-0.15

.jdesktop

-0.15

chet

-0.15

rys

-0.14

ikat

-0.14

ãģĴ

-0.14

emain

-0.14

ÃŃrk

-0.14

itles

-0.13

POSITIVE LOGITS

 Torrent

0.17

Ø³Ø§ÙĨÛĮ

0.15

Kop

0.15

 tÃŃm

0.14

amples

0.14

é³´

0.14

-LAST

0.14

éĶ

0.14

áº©m

0.14

.ribbon

0.14

Activations Density 0.133%

methods and strategies related to best practices in various contexts

No Comments

No Known Activations