INDEX

Explanations

negations and cautions regarding actions or recommendations

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

ux

-0.15

AMPL

-0.15

ssel

-0.14

.pivot

-0.14

izon

-0.14

tring

-0.14

rung

-0.14

Gos

-0.14

 somehow

-0.14

ubu

-0.14

POSITIVE LOGITS

 exceed

0.30

 EVER

0.28

 ever

0.26

 touch

0.23

 hesitate

0.22

 exceeds

0.22

 worry

0.22

 Ð¿ÑĢÐµÐ²ÑĭÑĪ

0.22

 exceeding

0.21

-ever

0.21

Activations Density 0.121%

negations and cautions regarding actions or recommendations

No Comments

No Known Activations