INDEX

Explanations

technical terms related to network communication and configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 dist

-0.16

roke

-0.15

patches

-0.15

-dismiss

-0.15

Ã¼z

-0.15

çĥ

-0.15

Semaphore

-0.14

antee

-0.14

erti

-0.14

 productive

-0.13

POSITIVE LOGITS

 tunnel

0.27

 tunnels

0.27

 encaps

0.26

 Tunnel

0.24

enc

0.21

Enc

0.21

tun

0.21

_tunnel

0.20

 shim

0.20

 header

0.18

Activations Density 0.004%

technical terms related to network communication and configurations

No Comments

No Known Activations