INDEX

Explanations

abbreviations or acronyms used in various contexts

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

ef

-0.24

-0.21

ec

-0.20

est

-0.20

eh

-0.19

ecom

-0.19

uci

-0.19

ir

-0.19

eb

-0.18

POSITIVE LOGITS

ollar

0.17

istrovstvÃŃ

0.17

yer

0.16

ansk

0.16

ubs

0.16

yntax

0.16

quo

0.15

yth

0.15

vor

0.15

unken

0.15

Activations Density 0.054%

abbreviations or acronyms used in various contexts

No Comments

No Known Activations