INDEX

Explanations

references to different groups of people or variations in opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Ð»Ð¾Ð½

-0.15

mlink

-0.15

JA

-0.14

logan

-0.14

icken

-0.14

 HÃ¼k

-0.14

thus

-0.14

lrt

-0.14

urus

-0.14

iversit

-0.14

POSITIVE LOGITS

 simply

0.21

ones

0.20

 others

0.19

 merely

0.18

 simplement

0.17

åīĩ

0.17

 Ð¿ÑĢÐ¾ÑģÑĤÐ¾

0.16

 naopak

0.16

 ê·¸ëĥ¥

0.16

åĪĻ

0.15

Activations Density 0.041%

references to different groups of people or variations in opinions

No Comments

No Known Activations