INDEX

Explanations

references to academic databases and their contents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 corre

-0.16

ijkstra

-0.15

adata

-0.15

 Lear

-0.14

enge

-0.14

dev

-0.14

 Shuffle

-0.14

å®Ĺ

-0.14

continuous

-0.14

ALLOC

-0.13

POSITIVE LOGITS

ibraries

0.21

(library

0.19

.library

0.19

indow

0.18

 librarian

0.17

oodle

0.17

é¦Ĩ

0.17

ynet

0.15

/resource

0.15

implify

0.15

Activations Density 0.035%

references to academic databases and their contents

No Comments

No Known Activations