INDEX

Explanations

instances of dining and food-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 bake

-0.20

 baking

-0.18

ekli

-0.17

 brewing

-0.16

 baked

-0.15

anol

-0.15

brew

-0.15

 brew

-0.15

 entry

-0.14

 cakes

-0.14

POSITIVE LOGITS

 forks

0.31

 fork

0.30

fork

0.30

 plate

0.29

 Fork

0.28

 plates

0.26

 Plate

0.25

Plate

0.24

 Plates

0.23

_plate

0.23

Activations Density 0.167%

instances of dining and food-related activities

No Comments

No Known Activations