INDEX

Explanations

keywords indicating numerical values or relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ICODE

-0.15

 Guth

-0.15

Ø¶ÙĬ

-0.15

ourg

-0.14

 gore

-0.14

eldon

-0.14

RenderingContext

-0.14

íĦ¸

-0.13

Relative

-0.13

ÎµÎ²

-0.13

POSITIVE LOGITS

å±Ģ

0.16

ju

0.14

-fired

0.14

uly

0.14

ê³Ħ

0.14

iasi

0.14

hots

0.13

sum

0.13

unas

0.13

ynom

0.13

Activations Density 0.001%

keywords indicating numerical values or relationships

No Comments

No Known Activations