INDEX

Explanations

references to historical figures and leadership roles in a cultural context

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

 Sabha

-0.15

 Victorian

-0.15

lah

-0.14

á»ĵi

-0.14

 Laure

-0.13

 svob

-0.13

 French

-0.13

ull

-0.13

POSITIVE LOGITS

Hou

0.20

 Tang

0.20

Han

0.20

 Jose

0.19

Han

0.18

éļ

0.18

 Yellow

0.18

Yellow

0.17

 Bian

0.17

 HÃłn

0.17

Activations Density 0.010%

references to historical figures and leadership roles in a cultural context

No Comments

No Known Activations