INDEX

Explanations

terms and variations related to "order" and "ordering."

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

peror

-0.17

porter

-0.15

kart

-0.15

scores

-0.15

ÑĢÐ°Ð·

-0.14

ãģĳãĤĮãģ°

-0.14

lush

-0.14

ropolis

-0.14

eki

-0.14

angu

-0.14

POSITIVE LOGITS

liness

0.28

lies

0.20

/order

0.20

ings

0.20

iffin

0.19

edList

0.19

ments

0.18

.Order

0.17

inals

0.16

-order

0.15

Activations Density 0.058%

terms and variations related to "order" and "ordering."

No Comments

No Known Activations

terms and variations related to "order" and "ordering."

No Comments

No Known Activations