INDEX

Explanations

terms related to aerospace engineering and testing

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

-0.16

 Alps

-0.15

 Peaks

-0.14

 Ratio

-0.14

 Alpine

-0.14

 aerial

-0.14

yers

-0.13

sessions

-0.13

 Marie

-0.13

 ownership

-0.13

POSITIVE LOGITS

 launch

0.28

 Launch

0.28

Launch

0.26

launch

0.23

 launches

0.23

pad

0.23

 rocket

0.23

 Vand

0.23

-launch

0.22

 booster

0.21

Activations Density 0.022%

terms related to aerospace engineering and testing

No Comments

No Known Activations