INDEX

Explanations

programming-related terms and structures, particularly concerning recommendation systems and data processing parameters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

muz

-0.17

ÑĨÐµÐ½ÑĤÑĢÐ°

-0.16

 SQLITE

-0.16

 "\↵

-0.15

vib

-0.14

menin

-0.14

 ÙħÙĥ

-0.14

Ã¼ck

-0.13

obe

-0.13

meni

-0.13

POSITIVE LOGITS

 surrounding

0.18

avras

0.16

 Surround

0.15

ÑĢÑĥÑģ

0.15

 buffered

0.14

 hood

0.14

 surround

0.14

aval

0.13

Activations Density 0.200%

programming-related terms and structures, particularly concerning recommendation systems and data processing parameters

No Comments

No Known Activations