INDEX

Explanations

references to humor and comedy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

enie

-0.18

ffen

-0.17

ed

-0.16

chl

-0.16

 humorous

-0.16

ivr

-0.16

holder

-0.16

edb

-0.15

POSITIVE LOGITS

bone

0.23

erals

0.19

 relief

0.19

 Relief

0.19

 bone

0.19

 Bone

0.17

ously

0.17

Bone

0.17

 Bones

0.17

bones

0.16

Activations Density 0.033%

references to humor and comedy

No Comments

No Known Activations