INDEX

Explanations

instances of the verb "is" indicating definitions or states of being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

urement

-0.17

uada

-0.16

odos

-0.15

Ã¡h

-0.15

eter

-0.15

vÃ¤

-0.14

ound

-0.14

ëĵĿ

-0.14

istine

-0.13

opause

-0.13

POSITIVE LOGITS

 Part

0.17

 dedicated

0.17

ë³µ

0.16

yll

0.15

Ð»Ð°Ð¼

0.15

ector

0.15

ombat

0.15

 dedicate

0.15

yz

0.14

 adapted

0.14

Activations Density 0.040%

instances of the verb "is" indicating definitions or states of being

No Comments

No Known Activations