INDEX

Explanations

phrases indicating substitution or replacement of concepts or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

azon

-0.17

vla

-0.16

ENCHMARK

-0.15

Rew

-0.14

agle

-0.14

Ð½ÑĮ

-0.14

agrid

-0.14

ovu

-0.14

uren

-0.14

 Bever

-0.14

POSITIVE LOGITS

 substitute

0.27

replace

0.27

 replace

0.26

 replacing

0.24

 replaces

0.23

 substit

0.22

 replacement

0.22

 Replace

0.21

 substitution

0.21

 replaced

0.20

Activations Density 0.097%

phrases indicating substitution or replacement of concepts or items

No Comments

No Known Activations