INDEX

Explanations

questions or inquiries in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 anything

-0.16

inis

-0.15

avic

-0.15

agina

-0.14

compat

-0.14

ÑģÑĤÑİ

-0.14

anything

-0.14

 Anything

-0.14

uce

-0.14

stuff

-0.14

POSITIVE LOGITS

do

0.22

 else

0.19

if

0.18

 About

0.17

 about

0.17

 follows

0.17

aston

0.17

we

0.17

 better

0.16

ĸī

0.15

Activations Density 0.045%

questions or inquiries in the text

No Comments

No Known Activations