INDEX

Explanations

references to online education and related entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olas

-0.18

ÙĦØ§Øª

-0.15

ÏĦÎ¹ÏĤ

-0.15

æº

-0.14

ÃŃr

-0.14

>manual

-0.14

airs

-0.14

 elim

-0.14

ensa

-0.14

 reprint

-0.14

POSITIVE LOGITS

 course

0.30

Ud

0.27

 courses

0.25

 Course

0.25

 Cour

0.24

edx

0.24

course

0.23

Course

0.23

-course

0.23

 Courses

0.22

Activations Density 0.071%

references to online education and related entities

No Comments

No Known Activations