INDEX

Explanations

references to product formats and preferences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 Represents

-0.17

ras

-0.15

zh

-0.14

 tastes

-0.14

ombine

-0.14

CHandle

-0.14

comm

-0.14

bere

-0.13

oke

-0.13

 addCriterion

-0.13

POSITIVE LOGITS

 lies

0.31

lie

0.29

is

0.26

çļĦæĺ¯

0.23

lies

0.23

Lie

0.22

 include

0.22

 besides

0.21

,is

0.21

 Lies

0.21

Activations Density 0.129%

references to product formats and preferences

No Comments

No Known Activations