INDEX

Explanations

terms related to active roles and participation in a community context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ActivityResult

-0.18

 activating

-0.17

eni

-0.16

Exiting

-0.16

éĲĺ

-0.16

 activate

-0.16

aires

-0.15

sworth

-0.15

 Vict

-0.15

arily

-0.15

POSITIVE LOGITS

/pass

0.35

-duty

0.32

 ingredient

0.28

 duty

0.27

/react

0.26

 ingredients

0.25

 Ingredients

0.23

 Ingredient

0.23

 Duty

0.21

 participation

0.21

Activations Density 0.027%

terms related to active roles and participation in a community context

No Comments

No Known Activations