INDEX

Explanations

references to correspondence and author details in academic articles

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

Not in Any Lists

No Comments

Negative Logits

iously

-0.15

WS

-0.14

Presenter

-0.14

 Greater

-0.14

erie

-0.14

otty

-0.14

Ã¡o

-0.14

à¸²à¸ĸ

-0.14

ch

-0.14

Î¯Î¿

-0.13

POSITIVE LOGITS

ãĥĨãĥ«

0.15

ungan

0.15

inspace

0.15

olta

0.15

esModule

0.15

abwe

0.14

 demokrat

0.14

/part

0.14

 ÐºÐ°ÑĦ

0.14

 kvin

0.14

Activations Density 0.011%

references to correspondence and author details in academic articles

No Comments

No Known Activations