INDEX

Explanations

key terms and phrases associated with performance metrics and disruptions related to music or entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 Blanch

-0.15

 Idle

-0.15

-0.14

 Graz

-0.14

fry

-0.14

Idle

-0.14

actus

-0.14

Gran

-0.14

PK

-0.14

 Klein

-0.14

POSITIVE LOGITS

nip

0.26

arp

0.24

ip

0.23

rip

0.23

ÑĢÐ¸Ð¿

0.23

zip

0.23

Ð¸Ð¿

0.22

lip

0.22

åŁ

0.22

ap

0.22

Activations Density 0.023%

key terms and phrases associated with performance metrics and disruptions related to music or entertainment

No Comments

No Known Activations