INDEX

Explanations

programming constructs involving variable declarations and loops

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

å²

-0.14

usta

-0.14

alse

-0.14

ëł

-0.13

 Garrett

-0.13

(EIF

-0.13

"~

-0.13

arrant

-0.13

ä¸²

-0.13

inand

-0.13

POSITIVE LOGITS

Loop

0.23

_loop

0.22

 Loop

0.21

Ctr

0.21

 loop

0.20

loop

0.20

-loop

0.18

{}

0.18

 loops

0.18

-counter

0.17

Activations Density 0.047%

programming constructs involving variable declarations and loops

No Comments

No Known Activations