INDEX

Explanations

proper names and terms from various languages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

wcs

-0.16

jq

-0.15

oueur

-0.15

wie

-0.15

rico

-0.14

eah

-0.14

eldon

-0.14

ebra

-0.14

mons

-0.14

ajs

-0.14

POSITIVE LOGITS

Uy

0.23

 poru

0.22

ipp

0.22

 ullam

0.22

 uyar

0.21

vaz

0.20

enn

0.20

uy

0.20

 peru

0.20

 penn

0.20

Activations Density 0.024%

proper names and terms from various languages

No Comments

No Known Activations