INDEX

Explanations

technical terms related to software architecture and design practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

omid

-0.18

 Perkins

-0.15

arpa

-0.14

 Persistent

-0.14

meer

-0.14

opa

-0.14

à¹Ģà¸

-0.14

orio

-0.14

oppins

-0.14

 perman

-0.14

POSITIVE LOGITS

icator

0.15

 Sole

0.14

 nÃ³

0.14

rete

0.14

Yue

0.14

achat

0.14

æ´

0.14

 Truy

0.14

unte

0.14

Activations Density 0.247%

technical terms related to software architecture and design practices

No Comments

No Known Activations