INDEX

Explanations

terms related to industrial processes and facilities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

-0.14

anine

-0.14

à¸Ĺà¸°

-0.13

ÑĢÐ°Ð²Ð¸

-0.13

ainty

-0.13

gag

-0.13

atra

-0.13

zel

-0.13

kli

-0.12

 Repair

-0.12

POSITIVE LOGITS

isay

0.16

é¡¿

0.15

éłĵ

0.14

 facility

0.13

ÏĦÎ¹ÏĤ

0.13

anford

0.13

 Norman

0.13

cale

0.13

.Marshal

0.13

æ½®

0.13

Activations Density 0.065%

terms related to industrial processes and facilities

No Known Activations