INDEX

Explanations

specific numerical or regiment-related information

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

kaar

-0.16

ecta

-0.16

adoo

-0.16

ccione

-0.15

acro

-0.15

.opend

-0.15

ivec

-0.15

kening

-0.15

Ł¥

-0.14

unkt

-0.14

POSITIVE LOGITS

 Volunteers

0.26

 volunteers

0.23

 Companies

0.22

 Volunteer

0.21

 Company

0.21

 volunteer

0.20

Inf

0.20

zou

0.19

Inf

0.19

 companies

0.18

Activations Density 0.015%

specific numerical or regiment-related information

No Comments

No Known Activations