INDEX

Explanations

mentions of locations and their characteristics or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Performed

-0.17

 stretched

-0.17

Mounted

-0.15

 Stored

-0.15

Ð¼ÐµÐ½Ð°

-0.15

Displayed

-0.15

uze

-0.15

 housed

-0.15

anga

-0.14

Stored

-0.14

POSITIVE LOGITS

 frequ

0.30

 visited

0.30

 occupied

0.28

visited

0.24

 accessed

0.22

 overrun

0.22

 inhabited

0.22

 carpet

0.22

 host

0.21

 hosting

0.21

Activations Density 0.375%

mentions of locations and their characteristics or conditions

No Comments

No Known Activations