INDEX

Explanations

references to food and catering services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 descending

-0.15

 Desc

-0.14

xCD

-0.14

 chocol

-0.14

 harmless

-0.14

 cogn

-0.14

Ø²ÙĬ

-0.14

kovi

-0.14

 Restoration

-0.14

 network

-0.13

POSITIVE LOGITS

 carry

0.35

 take

0.30

 grab

0.29

carry

0.28

 Carry

0.27

take

0.27

grab

0.26

 catering

0.26

 Take

0.26

 TAKE

0.25

Activations Density 0.116%

references to food and catering services

No Comments

No Known Activations