INDEX

Explanations

expressions related to illness or feeling unwell

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

asca

-0.17

ydÄ±

-0.17

à¸²à¸į

-0.16

egers

-0.15

ept

-0.15

quare

-0.15

hythm

-0.14

ÐµÐºÐ°ÑĢ

-0.14

ife

-0.14

phy

-0.14

POSITIVE LOGITS

ening

0.39

ened

0.34

ens

0.24

 sick

0.23

ly

0.22

lesh

0.21

bed

0.20

lier

0.20

bay

0.20

 Sick

0.20

Activations Density 0.011%

expressions related to illness or feeling unwell

No Comments

No Known Activations