INDEX

Explanations

references to educational guidelines and standards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

atti

-0.17

istrovstvÃŃ

-0.16

esda

-0.16

åł¡

-0.14

 çł

-0.13

BorderStyle

-0.13

ÑĥÐ¶Ð´

-0.13

uzu

-0.13

DataSet

-0.13

ints

-0.13

POSITIVE LOGITS

cur

0.43

cur

0.43

 curriculum

0.38

-cur

0.34

Cur

0.34

.cur

0.33

Cur

0.30

CUR

0.30

 Curriculum

0.29

 content

0.29

Activations Density 0.233%

references to educational guidelines and standards

No Comments

No Known Activations