INDEX

Explanations

expressions related to governmental or administrative actions and announcements

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

ismet

-0.16

arendra

-0.16

Sans

-0.15

abad

-0.15

 Sans

-0.14

Ã³nica

-0.14

sans

-0.14

ÉĻ

-0.14

sei

-0.14

hai

-0.14

POSITIVE LOGITS

 ÙĢ

0.20

.ps

0.20

Fay

0.19

awy

0.18

 keen

0.17

Diy

0.16

 Jordan

0.16

 axes

0.16

 Nile

0.16

 âĢ«

0.16

Activations Density 0.350%

expressions related to governmental or administrative actions and announcements

No Comments

No Known Activations