INDEX

Explanations

phrases that indicate a rephrasing or simplification of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

straints

-0.17

erable

-0.15

zan

-0.15

ÙħØŃ

-0.15

adius

-0.14

straint

-0.14

tring

-0.14

ogue

-0.14

 Telescope

-0.14

anou

-0.14

POSITIVE LOGITS

xi

0.15

âĸį

0.15

ph

0.15

arb

0.15

appers

0.14

isco

0.14

Ð»ÑİÐ´

0.14

MDB

0.14

phrase

0.14

ropa

0.14

Activations Density 0.167%

phrases that indicate a rephrasing or simplification of information

No Comments

No Known Activations