INDEX

Explanations

references to qualifications and professional backgrounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ð¾Ð¿ÑĢÐ¸

-0.15

è¼Ŀ

-0.15

ankan

-0.15

etus

-0.15

ÑīÐ¸

-0.15

olum

-0.14

à¹Ħà¸£

-0.13

Indexed

-0.13

 Ð¿Ð¾ÑģÑĤÑĥÐ¿

-0.13

aceut

-0.13

POSITIVE LOGITS

 expert

0.21

expert

0.18

 experts

0.18

.sdk

0.18

 expertise

0.17

 responseObject

0.15

 Expert

0.15

arges

0.15

er

0.14

à¹Ģà¸īà¸¥

0.14

Activations Density 0.191%

references to qualifications and professional backgrounds

No Comments

No Known Activations