INDEX

Explanations

proper nouns and names related to individuals and organizations 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 @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

yb

-0.17

ossier

-0.17

aleigh

-0.16

å±ĭ

-0.16

andest

-0.16

kÃ¼

-0.16

-transparent

-0.16

ÃŃses

-0.15

aspers

-0.15

Ã©ric

-0.15

POSITIVE LOGITS

owski

0.41

inski

0.32

ski

0.32

ows

0.30

sky

0.28

iew

0.27

kowski

0.25

cz

0.25

czy

0.25

ka

0.23

Activations Density 0.090%

proper nouns and names related to individuals and organizations in various contexts

No Comments

No Known Activations