INDEX

Explanations

phrases that highlight the concept of emphasis

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

-0.15

ÑĪÐµÐ²

-0.15

lement

-0.15

erc

-0.15

adero

-0.14

 æĤ

-0.14

site

-0.14

erver

-0.14

zin

-0.14

ingu

-0.14

POSITIVE LOGITS

uated

0.23

 placed

0.23

phasis

0.22

 emphasis

0.22

emphasis

0.21

æĶ¾åľ¨

0.19

placed

0.19

 accent

0.18

uating

0.18

 Placement

0.18

Activations Density 0.021%

phrases that highlight the concept of emphasis

No Comments

No Known Activations