INDEX

Explanations

words and phrases related to emotional responses and expressions of affection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ervo

-0.17

 Sesso

-0.15

azon

-0.14

uiten

-0.14

.energy

-0.14

ä¼Ļ

-0.13

serial

-0.13

agen

-0.13

YPE

-0.13

à¤łà¤¨

-0.13

POSITIVE LOGITS

 sadness

0.41

 sorrow

0.38

sad

0.38

Sad

0.34

Sad

0.34

 sentimental

0.33

 mourn

0.32

 tears

0.32

sad

0.32

 saddened

0.31

Activations Density 0.006%

words and phrases related to emotional responses and expressions of affection

No Comments

No Known Activations