INDEX

Explanations

occurrences of the word "By" indicating authorship or attribution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ð°Ñı

-0.17

lio

-0.17

ught

-0.15

vest

-0.15

à¤Ī

-0.15

èµ·æĿ¥

-0.14

stÅĻÃŃ

-0.14

hattan

-0.14

Ã¢n

-0.14

ston

-0.14

POSITIVE LOGITS

orthand

0.16

 admin

0.16

'gc

0.14

ernes

0.14

ington

0.14

ìĿ´ìŀĲ

0.14

Chance

0.14

 chance

0.14

 Unsigned

0.13

ÙĨÙĩ

0.13

Activations Density 0.018%

occurrences of the word "By" indicating authorship or attribution

No Comments

No Known Activations