INDEX

Explanations

coding-related terminology and properties in styles and animations

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

iske

-0.18

ataka

-0.15

.ld

-0.15

atus

-0.15

andon

-0.14

aukee

-0.14

avis

-0.14

imon

-0.14

pty

-0.14

onz

-0.14

POSITIVE LOGITS

AccessType

0.14

 FileAccess

0.14

é¢Ħè§Ī

0.14

_FALL

0.13

èİ

0.13

gnore

0.13

oce

0.13

=”

0.13

Decor

0.13

Activations Density 0.003%

coding-related terminology and properties in styles and animations

No Comments

No Known Activations