INDEX

Explanations

phrases reflecting programming concepts and integration within software applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 without

-0.17

gent

-0.15

 WITHOUT

-0.14

 quick

-0.14

without

-0.14

kiye

-0.14

[:]

-0.14

ć

-0.14

owel

-0.14

ison

-0.13

POSITIVE LOGITS

 manually

0.23

each

0.22

 ÐºÐ°Ð¶Ð´ÑĭÐ¹

0.21

 separate

0.21

 separately

0.21

 EACH

0.20

à¸¡à¸²à¸ģà¸¡à¸²à¸¢

0.20

ç¹ģ

0.20

 costly

0.20

 each

0.20

Activations Density 0.322%

phrases reflecting programming concepts and integration within software applications

No Comments

No Known Activations