INDEX

Explanations

numerical expressions related to quantities, prices, or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

reon

-0.13

avic

-0.13

 Scho

-0.13

à¸±à¹Ī

-0.13

 topp

-0.12

ouro

-0.12

afi

-0.12

enu

-0.12

çĶ±

-0.12

 conse

-0.12

POSITIVE LOGITS

at

0.84

 táº¡i

0.40

èĩ³å°ĳ

0.40

at

0.39

 á»Ł

0.39

_at

0.38

at

0.34

.at

0.32

 atIndex

0.32

bei

0.31

Activations Density 0.274%

numerical expressions related to quantities, prices, or measurements

No Comments

No Known Activations