INDEX

Explanations

references to defense-related systems and programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ãĥ¼ãĥĲ

-0.17

 Telecom

-0.17

 butterfly

-0.16

 Antar

-0.15

kad

-0.14

MethodInfo

-0.14

azz

-0.14

 TÃ©lÃ©

-0.14

dÃ¼

-0.14

 Bentley

-0.14

POSITIVE LOGITS

Insensitive

0.21

 Joint

0.20

 solic

0.18

Joint

0.18

 Lightweight

0.18

-Ray

0.17

 Solic

0.17

 Army

0.17

Ray

0.16

Ray

0.16

Activations Density 0.028%

references to defense-related systems and programs

No Comments

No Known Activations