INDEX

Explanations

references to hidden or less visible aspects of situations or processes

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

.nano

-0.17

adiens

-0.15

idan

-0.15

shal

-0.15

anky

-0.14

_VARIABLE

-0.14

ioc

-0.14

aybe

-0.14

iah

-0.14

uno

-0.14

POSITIVE LOGITS

 closed

0.42

 Closed

0.35

 scenes

0.33

closed

0.33

Closed

0.32

 CLOSED

0.31

 curtain

0.29

-scenes

0.29

-the

0.29

 curtains

0.28

Activations Density 0.024%

references to hidden or less visible aspects of situations or processes

No Comments

No Known Activations