INDEX

Explanations

numerical values and code structure indicative of programming or styling elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 seiz

-0.17

iyon

-0.15

à¸Ľà¸£à¸°

-0.15

ë²Ķ

-0.15

 Huck

-0.15

isco

-0.15

↵

-0.14

zed

-0.14

POSITIVE LOGITS

ICO

0.15

ipp

0.15

ico

0.15

ÐµÐ»ÑĮ

0.14

issance

0.14

 Wonder

0.13

à¹īà¸²à¸ķ

0.13

ect

0.13

essage

0.13

inen

0.13

Activations Density 0.020%

numerical values and code structure indicative of programming or styling elements

No Comments

No Known Activations