INDEX

Explanations

references to electrical concepts and terms related to electricity

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

 Thom

-0.17

 Birch

-0.17

expiration

-0.15

oyal

-0.15

ium

-0.15

ientes

-0.15

(_('

-0.15

à¥ĭà¤§

-0.14

ovy

-0.14

POSITIVE LOGITS

ro

0.31

ros

0.30

rol

0.28

rom

0.28

ron

0.26

rowave

0.25

rote

0.25

roph

0.24

row

0.23

rop

0.23

Activations Density 0.024%

references to electrical concepts and terms related to electricity

No Comments

No Known Activations