INDEX

Explanations

references to emotional experiences and responses

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

emotion

-0.17

 emoc

-0.17

 emotion

-0.17

reta

-0.16

 emotions

-0.16

okable

-0.16

inition

-0.15

pson

-0.14

 Emotional

-0.14

Ð»Ð¸ÑĨ

-0.14

POSITIVE LOGITS

ality

0.24

 intelligence

0.23

 roller

0.21

 charged

0.21

 blackmail

0.20

 Intelligence

0.20

 attachment

0.20

 regulation

0.19

 Roller

0.19

ÑĨÐ¸Ð¾Ð½Ð°Ð»ÑĮ

0.18

Activations Density 0.023%

references to emotional experiences and responses

No Comments

No Known Activations