INDEX

Explanations

phrases expressing regret or reflection on past decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

sville

-0.16

(éĩĳ

-0.14

 BaÄŁ

-0.14

lessly

-0.14

imest

-0.14

leen

-0.14

intree

-0.14

.fil

-0.14

roker

-0.14

alars

-0.14

POSITIVE LOGITS

iki

0.15

 fault

0.15

 principle

0.14

ackbar

0.14

ÐµÑĢÐ¸Ð½

0.14

heim

0.14

Backdrop

0.14

Facade

0.14

uÃ©

0.14

Activations Density 0.024%

phrases expressing regret or reflection on past decisions

No Comments

No Known Activations