INDEX

Explanations

information related to academic databases and research articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

agma

-0.15

Ä±t

-0.14

ckeditor

-0.14

ikh

-0.14

Strict

-0.14

ä¿

-0.14

etter

-0.13

ãĥ³ãĤ¹

-0.13

.Listener

-0.13

 interchange

-0.13

POSITIVE LOGITS

Lib

0.26

lib

0.26

 Summon

0.25

 libr

0.25

 librarian

0.23

Summon

0.22

-lib

0.22

Zot

0.22

.lib

0.20

 circulation

0.20

Activations Density 0.018%

information related to academic databases and research articles

No Comments

No Known Activations