INDEX

Explanations

mathematical operations and symbols in equations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

oux

-0.17

ieux

-0.15

_exists

-0.15

mos

-0.15

Ð»Ð¸ÑĤ

-0.14

ambre

-0.14

ALAR

-0.14

infeld

-0.13

nici

-0.13

Kup

-0.13

POSITIVE LOGITS

cdot

0.43

 âĢ¢

0.34

times

0.32

 times

0.32

âĢ¢

0.31

.âĢ¢

0.26

 Times

0.26

 Â·

0.25

Â·

0.25

_times

0.24

Activations Density 0.135%

mathematical operations and symbols in equations

No Comments

No Known Activations