INDEX

Explanations

numerical and possibly coded information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ylvania

-0.17

fur

-0.17

upe

-0.15

ë´ī

-0.15

ConnectionString

-0.14

fur

-0.14

utations

-0.14

contre

-0.14

uye

-0.14

Ð·Ð°Ð±

-0.14

POSITIVE LOGITS

ignon

0.14

sher

0.14

bridge

0.14

_HW

0.14

 adipiscing

0.13

omed

0.13

ffset

0.13

.FileReader

0.13

WND

0.13

agna

0.13

Activations Density 0.012%

numerical and possibly coded information

No Comments

No Known Activations