INDEX

Explanations

instances of the word "or."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/or

-0.24

swer

-0.18

Ð½Ñİ

-0.17

venir

-0.17

emas

-0.16

enty

-0.16

ì¹

-0.15

oria

-0.15

/OR

-0.15

POSITIVE LOGITS

ourke

0.29

 Bust

0.28

ignal

0.25

 bust

0.25

anged

0.24

yx

0.23

anging

0.22

acles

0.21

ifice

0.21

lando

0.21

Activations Density 0.141%

instances of the word "or."

No Comments

No Known Activations