INDEX

Explanations

content related to numerical or chronological details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 ÑĤÐ²Ð°

-0.17

 Ð¾Ð³ÑĢÐ°

-0.17

 Ð±ÑĥÐ¼Ð°

-0.17

 herself

-0.17

ê¸ī

-0.16

 Ø§ÙĦØ±ÙħØ²ÙĬØ©

-0.15

 ÑĤÐ°Ð±Ð»Ð¸

-0.14

ENCIES

-0.14

paque

-0.14

 )↵↵↵↵↵↵↵↵

-0.14

POSITIVE LOGITS

 ÑĢÐ°Ð±Ð¾ÑĤ

0.30

 worked

0.29

worked

0.28

 ÑĥÑĩÐ°

0.25

 ÑĢÐ°Ð±Ð¾ÑĤÐ°ÐµÑĤ

0.24

 Ð°ÐºÑĤÐ¸Ð²Ð½Ð¾

0.23

 ÑĥÑĩÐ°Ñģ

0.23

 trabaj

0.22

 lavor

0.22

 Ð¾ÐºÐ¾Ð½

0.21

Activations Density 0.020%

content related to numerical or chronological details

No Comments

No Known Activations