INDEX

Explanations

phrases related to warnings and cautions regarding safety and security

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

.scalablytyped

-0.16

ligt

-0.15

atik

-0.15

ë¦¼

-0.15

iola

-0.14

eft

-0.14

á»Ĳ

-0.14

wat

-0.14

 Inherits

-0.14

orgh

-0.14

POSITIVE LOGITS

 shortcut

0.16

use

0.16

DIY

0.15

 resort

0.15

 shortcuts

0.14

 forbidden

0.14

 retal

0.14

uden

0.14

 uses

0.14

 illegal

0.14

Activations Density 0.274%

phrases related to warnings and cautions regarding safety and security

No Comments

No Known Activations