INDEX

Explanations

programming-related keywords and commands in a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 FALL

-0.14

.mit

-0.14

usat

-0.14

LSB

-0.14

 Rowe

-0.14

 Colomb

-0.14

ewan

-0.13

attr

-0.13

_fh

-0.13

ahead

-0.13

POSITIVE LOGITS

Win

0.29

 Windows

0.27

WIN

0.25

Win

0.25

win

0.24

.exe

0.23

.bat

0.23

.win

0.23

_win

0.23

Windows

0.22

Activations Density 0.160%

programming-related keywords and commands in a technical context

No Comments

No Known Activations