INDEX

Explanations

references to people and their relationships, particularly focusing on notable individuals and their accomplishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

æķĻ

-0.17

elper

-0.14

ÐºÐ°ÑĤ

-0.14

uan

-0.14

nh

-0.14

&&!

-0.14

 Collapse

-0.14

æŁ´

-0.13

ABI

-0.13

uss

-0.13

POSITIVE LOGITS

(C

0.28

(CC

0.24

CW

0.22

CV

0.19

CY

0.19

CC

0.19

.CV

0.19

 Ð¦

0.18

CU

0.18

,C

0.17

Activations Density 0.216%

references to people and their relationships, particularly focusing on notable individuals and their accomplishments

No Comments

No Known Activations