INDEX

Explanations

phrases related to health and well-being

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

yas

-0.16

-handle

-0.16

health

-0.16

avez

-0.16

insky

-0.15

verts

-0.15

ifndef

-0.15

ikt

-0.14

oras

-0.14

-ÑĤÐ°ÐºÐ¸

-0.14

POSITIVE LOGITS

iness

0.33

iest

0.33

ily

0.30

fulness

0.27

 care

0.27

ful

0.25

ier

0.24

-care

0.24

fully

0.23

 Care

0.22

Activations Density 0.046%

phrases related to health and well-being

No Comments

No Known Activations