INDEX

Explanations

questions and phrases indicating uncertainty or wonder

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

Hey

-0.16

ecom

-0.15

agen

-0.14

culus

-0.14

uce

-0.14

igit

-0.14

obook

-0.14

Ø²ÛĮ

-0.14

.Net

-0.14

ìĦ¸

-0.14

POSITIVE LOGITS

 yeah

0.17

 Multiply

0.16

Ð»Ð°Ð¼Ð¸

0.15

berger

0.15

 Ä°ÅŁte

0.15

ike

0.15

 Morav

0.15

Multiply

0.15

 chances

0.15

 Yeah

0.15

Activations Density 0.067%

questions and phrases indicating uncertainty or wonder

No Comments

No Known Activations