INDEX

Explanations

references to governmental or administrative contexts

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

(strpos

-0.16

zon

-0.14

zend

-0.14

ija

-0.14

azel

-0.14

efs

-0.14

angl

-0.14

å¸Į

-0.14

Attrs

-0.14

Vault

-0.14

POSITIVE LOGITS

Sax

0.34

 Lower

0.30

 Brand

0.30

Bad

0.29

Rh

0.28

 Upper

0.27

 Sach

0.27

Lower

0.26

Brand

0.25

Bad

0.24

Activations Density 0.050%

references to governmental or administrative contexts

No Comments

No Known Activations