INDEX

Explanations

references to individuals or organizations

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

elic

-0.20

ahi

-0.19

ey

-0.18

aisy

-0.17

eci

-0.17

ain

-0.17

-0.16

airy

-0.16

enu

-0.16

POSITIVE LOGITS

uer

0.33

ÃŁ

0.30

cken

0.29

chsel

0.28

iÃŁ

0.27

ÃŁen

0.27

chts

0.27

ichen

0.27

chter

0.26

ifen

0.26

Activations Density 0.027%

references to individuals or organizations

No Comments

No Known Activations