INDEX

Explanations

references to geographical locations and historical contexts

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

ãĥ³ãĤ¹

-0.13

otify

-0.13

bia

-0.13

_FAR

-0.13

 trá»«

-0.12

ãĥ¼ãĥĢ

-0.12

.isnan

-0.12

retval

-0.12

afia

-0.12

POSITIVE LOGITS

 press

0.35

sul

0.34

 nella

0.31

 nell

0.31

nei

0.29

nel

0.28

sui

0.28

 within

0.28

 sulla

0.28

 nelle

0.27

Activations Density 0.017%

references to geographical locations and historical contexts

No Comments

No Known Activations