INDEX

Explanations

numerical data and references in scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

fone

-0.15

suy

-0.14

 Cannon

-0.14

Hog

-0.14

imp

-0.14

 datum

-0.14

objs

-0.14

bil

-0.14

kker

-0.14

ic

-0.13

POSITIVE LOGITS

/slick

0.17

ante

0.16

etta

0.16

ardo

0.15

PIO

0.15

ahat

0.15

 stav

0.15

anke

0.14

 TRACE

0.14

Ø¡

0.14

Activations Density 0.025%

numerical data and references in scientific contexts

No Comments

No Known Activations