INDEX

Explanations

the preposition "in" and its variations based on context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ä¸ĭåİ»

-0.15

elters

-0.15

urve

-0.14

ignite

-0.14

prite

-0.13

 Below

-0.13

 $('#'

-0.13

Ã¼ss

-0.12

PERTY

-0.12

urs

-0.12

POSITIVE LOGITS

 store

0.31

tow

0.30

 sight

0.29

 play

0.27

 reserve

0.26

 there

0.25

 mind

0.25

 abundance

0.22

play

0.21

store

0.21

Activations Density 0.109%

the preposition "in" and its variations based on context

No Comments

No Known Activations