INDEX

Explanations

phrases indicative of personal needs and desires

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

åĿĬ

-0.20

onda

-0.18

onder

-0.17

ervas

-0.17

eldon

-0.16

dana

-0.16

çĽĹ

-0.16

imple

-0.15

 Intent

-0.15

oba

-0.15

POSITIVE LOGITS

 barg

0.26

 deserve

0.23

 deserves

0.20

 seek

0.19

 desire

0.18

 Barg

0.18

æīĢ

0.18

 bargain

0.17

 require

0.16

cov

0.16

Activations Density 0.052%

phrases indicative of personal needs and desires

No Comments

No Known Activations