INDEX

Explanations

phrases related to healthcare and treatment methodologies

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

 angel

-0.15

robat

-0.15

 Minority

-0.15

 èİ

-0.15

 Ð¿ÑĢÐ¸Ð±

-0.14

cker

-0.14

oq

-0.14

ecut

-0.14

 Patch

-0.14

angel

-0.13

POSITIVE LOGITS

cov

0.24

obs

0.22

 intercept

0.21

 Intercept

0.20

 latent

0.20

Cov

0.20

 observations

0.20

 outcome

0.19

 nuisance

0.19

 observed

0.19

Activations Density 0.020%

phrases related to healthcare and treatment methodologies

No Comments

No Known Activations