INDEX

Explanations

terms related to palliative care and support for patients

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

remen

-0.15

 æ¬

-0.15

GIT

-0.14

uts

-0.14

utsch

-0.14

_HI

-0.14

.escape

-0.14

ustain

-0.14

pora

-0.14

 Stability

-0.14

POSITIVE LOGITS

 pall

0.31

 dying

0.27

 terminal

0.27

 Pall

0.27

 Terminal

0.26

 hosp

0.25

 Death

0.23

 death

0.23

 Hosp

0.23

Terminal

0.22

Activations Density 0.060%

terms related to palliative care and support for patients

No Comments

No Known Activations