INDEX

Explanations

references to specific products, brands, or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ayne

-0.19

 Prospect

-0.15

[^

-0.13

anko

-0.13

olid

-0.13

Ð±ÑĥÐ´ÑĮ

-0.13

airo

-0.13

tir

-0.13

ichern

-0.13

ä»£

-0.13

POSITIVE LOGITS

out

0.57

(out

0.30

 around

0.29

*out

0.28

åĩº

0.28

Out

0.28

out

0.28

_out

0.27

=out

0.27

-out

0.26

Activations Density 0.206%

references to specific products, brands, or items

No Comments

No Known Activations