INDEX

Explanations

instances of web addresses or URLs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

eno

-0.16

auen

-0.15

undy

-0.15

asio

-0.15

 tissue

-0.14

rello

-0.14

 Kingdom

-0.14

otty

-0.14

orical

-0.14

POSITIVE LOGITS

Teh

0.15

'icon

0.14

arel

0.14

hra

0.13

.omg

0.13

/respond

0.13

asaki

0.13

ubl

0.13

rap

0.13

 standards

0.13

Activations Density 0.013%

instances of web addresses or URLs

No Comments

No Known Activations