INDEX

Explanations

references to specific components and features of products or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ader

-0.16

ffa

-0.14

arkin

-0.14

*,

-0.14

harma

-0.14

pa

-0.14

 Halk

-0.14

(from

-0.14

oder

-0.13

POSITIVE LOGITS

way

0.36

WAY

0.23

 again

0.22

way

0.22

Way

0.21

 latter

0.20

 step

0.20

 alone

0.20

.way

0.20

WAY

0.19

Activations Density 0.133%

references to specific components and features of products or systems

No Comments

No Known Activations