INDEX

Explanations

statistical data and numerical information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

chet

-0.16

arda

-0.15

 attention

-0.15

xdd

-0.14

mong

-0.14

à¹īà¸²à¸ĩ

-0.13

íĻľ

-0.13

hton

-0.13

hora

-0.13

POSITIVE LOGITS

azu

0.15

forman

0.14

é¡ĺãģĦ

0.14

etest

0.14

 Innoc

0.14

-ci

0.14

Ø¡

0.14

pace

0.14

Ð¸Ð½Ð¾Ðº

0.14

 testCase

0.13

Activations Density 0.205%

statistical data and numerical information

No Comments

No Known Activations