INDEX

Explanations

coding-related terminology and programming 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

ihan

-0.16

otton

-0.16

ÎŃÏģÎµÎ¹

-0.16

 tail

-0.15

 itself

-0.14

irse

-0.14

ilden

-0.14

 rearr

-0.14

 hizo

-0.14

 pudo

-0.14

POSITIVE LOGITS

 yourself

0.32

ÐµÑĪÑĮ

0.26

 shalt

0.24

Ð¸ÑĤÐµ

0.23

 Yourself

0.23

tu

0.23

 kannst

0.22

 yourselves

0.22

 Ð¼Ð¾Ð¶ÐµÑĤÐµ

0.21

ites

0.21

Activations Density 0.055%

coding-related terminology and programming elements

No Comments

No Known Activations