INDEX

Explanations

URLs and XML-related content

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

izon

-0.16

-0.15

éĸĵãģ«

-0.15

ounding

-0.15

yan

-0.14

 Baker

-0.14

apolis

-0.14

ilingual

-0.13

opsis

-0.13

fos

-0.13

POSITIVE LOGITS

schemas

0.19

å®ĭä½ĵ

0.15

wash

0.14

 cazzo

0.13

Enumer

0.13

,LOCATION

0.13

.scalablytyped

0.13

å»ł

0.13

vetica

0.13

 bordel

0.13

Activations Density 0.024%

URLs and XML-related content

No Comments

No Known Activations