INDEX

Explanations

references to user data or publication information in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

rome

-0.16

urat

-0.14

aras

-0.14

.gs

-0.13

/WebAPI

-0.13

exc

-0.13

-dist

-0.13

-fontawesome

-0.13

cac

-0.13

POSITIVE LOGITS

PELL

0.16

ombo

0.14

 Academy

0.14

losures

0.14

 Ipsum

0.14

listing

0.14

ãĤ¹ãĥĨãĤ£

0.14

á»£i

0.14

è§Ĵ

0.13

á»Ļng

0.13

Activations Density 0.038%

references to user data or publication information in a programming context

No Comments

No Known Activations