INDEX

Explanations

programming-related terms and structures in coding contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

otas

-0.15

ÑĳÑĢ

-0.14

 ç¥

-0.14

amus

-0.13

 bÃ¼y

-0.13

allon

-0.13

ultiply

-0.13

aat

-0.13

licative

-0.13

Ø±Ø¯

-0.13

POSITIVE LOGITS

ervoir

0.15

nos

0.15

iny

0.15

usher

0.14

 ring

0.14

sy

0.14

ere

0.14

taj

0.14

 condom

0.14

put

0.14

Activations Density 0.034%

programming-related terms and structures in coding contexts

No Comments

No Known Activations