INDEX

Explanations

words related to academic and professional titles, particularly those denoting professors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

fang

-0.18

Äįek

-0.18

bara

-0.16

ques

-0.16

eric

-0.15

à¤¦à¤®

-0.15

Enlarge

-0.15

cznie

-0.15

è¹

-0.15

ATUS

-0.14

POSITIVE LOGITS

essed

0.38

anity

0.35

essions

0.33

usion

0.32

ess

0.32

iciency

0.31

iciencies

0.30

ession

0.29

use

0.29

esso

0.29

Activations Density 0.014%

words related to academic and professional titles, particularly those denoting professors

No Comments

No Known Activations