INDEX

Explanations

references to nutrition or nutritious foods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

esty

-0.19

hem

-0.17

aign

-0.16

haf

-0.15

stral

-0.15

ãĥ¼ãĥŃ

-0.15

hood

-0.15

yers

-0.15

otron

-0.14

POSITIVE LOGITS

ritional

0.33

meg

0.32

ty

0.30

rients

0.24

ters

0.24

rition

0.23

rient

0.22

shell

0.22

job

0.22

ted

0.21

Activations Density 0.007%

references to nutrition or nutritious foods

No Comments

No Known Activations