INDEX

Explanations

instances of a specific character or symbol often used in text

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

oring

-0.17

abe

-0.17

unt

-0.16

itive

-0.16

aret

-0.16

cing

-0.15

alt

-0.15

avers

-0.15

Ã¼lt

-0.15

odos

-0.14

POSITIVE LOGITS

Ð½ÑĨÐ¸ÐºÐ»Ð¾Ð¿ÐµÐ´

0.22

isko

0.21

Ð²Ð¾Ð»

0.17

Ð»

0.16

wart

0.16

olian

0.16

rm

0.16

lemen

0.15

umed

0.15

mission

0.15

Activations Density 0.004%

instances of a specific character or symbol often used in text

No Comments

No Known Activations