INDEX

Explanations

formatting elements and mathematical expressions in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

tn

-0.15

anik

-0.15

ewise

-0.14

ãģİ

-0.14

-tooltip

-0.14

è´¸

-0.14

ÑĨÐµÐ¹

-0.13

avia

-0.13

 FileStream

-0.13

iphy

-0.13

POSITIVE LOGITS

skip

0.21

fil

0.18

à¸Ħ

0.15

iola

0.14

break

0.14

 kern

0.14

 ince

0.14

ÑĪÐºÐ¸

0.14

$#

0.14

 advancing

0.14

Activations Density 0.046%

formatting elements and mathematical expressions in text

No Comments

No Known Activations