INDEX

Explanations

the word "all" and variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

inded

-0.16

edom

-0.14

eÃ§

-0.14

kiye

-0.14

-BEGIN

-0.14

beck

-0.14

QUIRE

-0.13

ä½Ĩ

-0.13

rone

-0.13

:"-

-0.13

POSITIVE LOGITS

 sorts

0.30

 kinds

0.28

iteration

0.24

sort

0.23

 those

0.22

iterations

0.22

 SORT

0.22

 that

0.22

igators

0.21

 KIND

0.21

Activations Density 0.061%

the word "all" and variations of it

No Comments

No Known Activations