INDEX

Explanations

references to diverse elements, particularly those related to quantities, numbers, and varying concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ÐºÐ¾ÑģÑĤÐ¸

-0.18

 bidi

-0.17

Ã¤Ã¤

-0.16

Ã©mu

-0.16

ÑĥÐºÐ¸

-0.15

ije

-0.15

emu

-0.15

Ð»ÐµÐ½Ð¸Ñİ

-0.15

allee

-0.15

nÃŃmu

-0.15

POSITIVE LOGITS

Ð´Ð°Ñħ

0.39

ÐºÐ°Ñħ

0.37

ÑĤÐ°Ñħ

0.37

ÐµÑĢÐ°Ñħ

0.37

ÑıÑħ

0.35

ÑĨÐ¸ÑıÑħ

0.35

ÑĢÐ°Ñħ

0.35

Ð°Ñħ

0.35

Ð½Ð°Ñħ

0.35

Ð¼Ð°Ñħ

0.34

Activations Density 0.033%

references to diverse elements, particularly those related to quantities, numbers, and varying concepts

No Comments

No Known Activations