INDEX

Explanations

references to direct effects or impacts on various subjects or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 direct

-0.23

 Direct

-0.23

Direct

-0.23

_direct

-0.21

 Ð´Ð¸ÑĢ

-0.20

 dire

-0.20

richt

-0.20

 diret

-0.20

_DIRECT

-0.20

direct

-0.20

POSITIVE LOGITS

ivity

0.30

ives

0.24

ivities

0.22

IVITY

0.21

 proportional

0.20

rices

0.20

eur

0.20

 doÄŁru

0.20

 ÐºÐ¸ÑĪ

0.19

oire

0.17

Activations Density 0.032%

references to direct effects or impacts on various subjects or situations

No Comments

No Known Activations